{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "### Dataset Loading and testing" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "ExecuteTime": { "end_time": "2018-10-10T05:46:44.554177Z", "start_time": "2018-10-10T05:46:44.334529Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The autoreload extension is already loaded. To reload it, use:\n", " %reload_ext autoreload\n", "Dataset has been loaded\n", "x-train (60000, 8)\n", "x-test (40000, 8)\n", "y-train (60000, 256)\n", "y-test (40000, 256)\n" ] } ], "source": [ "%load_ext autoreload\n", "%autoreload 2\n", "\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import h5py\n", "from sklearn.model_selection import train_test_split\n", "#import jtplot submodule from jupyterthemes\n", "from jupyterthemes import jtplot\n", "#currently installed theme will be used to\n", "#set plot style if no arguments provided\n", "jtplot.style()\n", "\n", "#now load this dataset \n", "h5f = h5py.File('./datasets/s8_sio2tio2_v2.h5','r')\n", "X = h5f['sizes'][:]\n", "Y = h5f['spectrum'][:]\n", "\n", "\n", "\n", "\n", "#get the ranges of the loaded data\n", "num_layers = X.shape[1]\n", "num_lpoints = Y.shape[1]\n", "size_max = np.amax(X)\n", "size_min = np.amin(X)\n", "size_av = 0.5*(size_max + size_min)\n", "\n", "#this information is not given in the dataset\n", "lam_min = 300\n", "lam_max = 1200\n", "lams = np.linspace(lam_min, lam_max, num_lpoints)\n", "\n", "# X = np.expand_dims(X, 1)\n", "# #X = np.expand_dims(X, 3)\n", "# Y = np.expand_dims(Y, 1)\n", "# #Y = np.expand_dims(Y, 3)\n", "\n", "\n", "\n", "\n", "\n", "\n", "#create a train - test split of the dataset\n", "x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.4, random_state=42)\n", "\n", "# normalize inputs \n", "x_train = (x_train - 50)/20 \n", "x_test = (x_test - 50)/20 \n", "\n", "print(\"Dataset has been loaded\")\n", "print(\"x-train\", x_train.shape)\n", "print(\"x-test \", x_test.shape)\n", "print(\"y-train\", y_train.shape)\n", "print(\"y-test \", y_test.shape)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Model Development" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-27T05:04:59.838628Z", "start_time": "2018-09-27T05:04:59.808085Z" } }, "outputs": [], "source": [ "from keras.utils import to_channels_first\n", "x_train = to_channels_first(x_train)\n" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "ExecuteTime": { "end_time": "2018-10-10T05:46:57.389010Z", "start_time": "2018-10-10T05:46:57.345387Z" }, "code_folding": [ 48, 65, 84, 123, 169, 234 ] }, "outputs": [], "source": [ "from keras import backend as K\n", "from keras.models import Sequential, Model\n", "from keras.layers import Dense, Dropout, Reshape, UpSampling1D, Conv1D, Flatten, Activation\n", "from keras.utils import np_utils, multi_gpu_model\n", "from keras.regularizers import l2\n", "from keras.wrappers.scikit_learn import KerasRegressor\n", "from keras.optimizers import Adam\n", "from keras.layers.normalization import BatchNormalization\n", "from keras.layers import PReLU\n", "\n", "\n", "from sklearn.model_selection import cross_val_score, KFold\n", "from sklearn.preprocessing import StandardScaler\n", "from sklearn.pipeline import Pipeline\n", "\n", "num_gpus = 2\n", "gpu_list = [\"gpu(%d)\" % i for i in range(num_gpus)]\n", "\n", "\n", "\n", "#define various models here\n", "#naive percentage loss\n", "def size_percent_loss(y_true, y_pred):\n", " y_true_a = 0.5*y_true*(size_max - size_min) + size_av\n", " y_pred_a = 0.5*y_pred*(size_max - size_min) + size_av\n", " y_err = np.abs(y_true_a - y_pred_a)/y_true_a\n", " y_err_f = K.flatten(y_err)\n", " return K.sum(y_err_f)\n", "\n", "#naive percentage loss\n", "def naive_percent_loss(y_true, y_pred):\n", " y_err = np.abs(y_true - y_pred)/y_true\n", " y_err_f = K.flatten(y_err)\n", " return K.sum(y_err_f)\n", "\n", "\n", "#function to test performance on testset \n", "def calc_mre(y_true, y_pred):\n", " y_err = 100*np.abs(y_true - y_pred)/y_true\n", " return np.mean(y_err)\n", "\n", "#function to test performance on testset \n", "def calc_mre_K(y_true, y_pred):\n", " y_err = 100*np.abs(y_true - y_pred)/y_true\n", " return K.mean(y_err)\n", "\n", "\n", "\n", "def naiveploss_mgpu_model():\n", " # create model\n", " model = Sequential()\n", " model = multi_gpu_model(model, gpus=num_gpus)\n", " model.add(Dense(250, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='first' ))\n", " model.add(Dense(250, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='second' ))\n", " model.add(Dense(250, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='third' ))\n", " model.add(Dense(250, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='fourth' ))\n", " model.add(Dense(250, kernel_initializer='normal', name='last'))\n", " # Compile model\n", " model.compile(loss=naive_percent_loss, optimizer='adam', context = gpu_list)\n", " return model\n", "\n", "def naiveploss_model():\n", " # create model\n", " model = Sequential()\n", " model.add(Dense(256, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='first' ))\n", " model.add(Dense(256, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='second' ))\n", " model.add(Dense(256, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='third' ))\n", " model.add(Dense(256, input_dim=x_train.shape[1], kernel_initializer='normal', activation='relu', \n", " name='fourth' ))\n", " model.add(Dense(256, kernel_initializer='normal', name='last'))\n", " # Compile model\n", " model.compile(loss=naive_percent_loss, optimizer='adam', metrics=['accuracy'])\n", " return model\n", "\n", "import timeit\n", "#here we must have a function that calls the training routine n times and then gives avg and stddev \n", "# of the resulting figures\n", "def net_performance(modelfunc, num_trials=3, batch_size=32, num_epochs=200, num_gpus=2):\n", " models = []\n", " train_err = np.ones(num_trials)\n", " test_err = np.ones(num_trials)\n", " val_err = np.ones(num_trials)\n", " train_time = np.ones(num_trials)\n", " for tnum in np.arange(num_trials):\n", " print(\"iteration: \" + str(tnum + 1))\n", " model_curr = modelfunc()\n", " x_t, x_v, y_t, y_v = train_test_split(x_train, y_train, test_size=0.2, random_state=42)\n", " start_time = timeit.default_timer()\n", " history = model_curr.fit(x_t, y_t,\n", " batch_size=batch_size*num_gpus,\n", " epochs=num_epochs, \n", " verbose=1,\n", " validation_data=(x_v, y_v))\n", " train_time[tnum] = timeit.default_timer() - start_time\n", " models.append(model_curr)\n", " train_err[tnum] = (100.0/num_lpoints)*history.history['loss'][-1]/(batch_size*num_gpus)\n", " val_err[tnum] = (100.0/num_lpoints)*history.history['val_loss'][-1]/(batch_size*num_gpus)\n", " test_err[tnum] = calc_mre(y_test, models[tnum].predict(x_test))\n", " return train_err, val_err, test_err, train_time\n", "\n", "#staging area for new models \n", "def plot_training_history(history, factor):\n", " loss, val_loss = history.history['loss'], history.history['val_loss']\n", " loss = np.asarray(loss)/(factor)\n", " val_loss = np.asarray(val_loss)/(factor)\n", " epochs = len(loss)\n", " \n", " fig, axs = plt.subplots(1,1, figsize=(5,2.5))\n", " axs.semilogy(np.arange(1, epochs + 1), loss, label='Train error')\n", " axs.semilogy(np.arange(1, epochs + 1), val_loss, label='Test error')\n", " axs.set_xlabel('Epoch number')\n", " #axs.set_ylim((0.4, 3))\n", " axs.set_xlim(left=1)\n", "# plt.yticks(np.array([0.1, 0.2, 0.3, 0.4, 0.5, 0.75, 1.0, 1.5, 2]), \n", "# ('0.1', '0.2', '0.3', '0.4', '0.5', '0.75', '1.0', '1.5', '2'))\n", " plt.yticks(np.array([0.5, 0.75, 1.0, 1.5, 2]), \n", " ('0.5', '0.75', '1.0', '1.5', '2'))\n", " axs.set_ylabel('MRE (%)')\n", " axs.legend(loc=\"best\")\n", " fig.savefig(\"foo2.pdf\", bbox_inches='tight')\n", "\n", "from keras.utils import to_channels_first\n", "\n", "def conv1d_lkyrelu():\n", " \n", " #gpu_list = [\"gpu(%d)\" % i for i in range(num_gpus)]\n", " \n", " # create model\n", " model = Sequential()\n", " \n", " model.add(Dense(256, input_dim=8, kernel_initializer='normal', \n", " name='first'))\n", " #model.add(BatchNormalization())\n", " #model.add(Activation('relu')) \n", " model.add(PReLU(alpha_initializer='zeros', alpha_regularizer=None))\n", " \n", " model.add(Reshape((4, 64)))\n", " model.add(UpSampling1D(size=2))\n", " \n", " \n", " model.add(Conv1D(filters=64, kernel_size=3, strides=1, padding='same', \n", " dilation_rate=1, \n", " kernel_initializer='normal'))\n", " model.add(PReLU(alpha_initializer='zeros', alpha_regularizer=None))\n", " \n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(PReLU(alpha_initializer='zeros', alpha_regularizer=None))\n", " \n", " model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", " dilation_rate=1, \n", " kernel_initializer='normal'))\n", " model.add(PReLU(alpha_initializer='zeros', alpha_regularizer=None))\n", "\n", "\n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(PReLU(alpha_initializer='ones', alpha_regularizer=None)) \n", " \n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(PReLU(alpha_initializer='ones', alpha_regularizer=None)) \n", " \n", " model.add(Flatten())\n", " # Compile model\n", " model.compile(loss=naive_percent_loss, optimizer='adam', metrics=[calc_mre_K])\n", " return model\n", " \n", " \n", "def conv1d_model_bnorm():\n", " \n", " #gpu_list = [\"gpu(%d)\" % i for i in range(num_gpus)]\n", " \n", " # create model\n", " model = Sequential()\n", " \n", " model.add(Dense(256, input_dim=8, kernel_initializer='normal', \n", " name='first' ))\n", " #model.add(BatchNormalization())\n", " model.add(Activation('relu'))\n", " #model.add(Dropout(0.2))\n", " \n", " model.add(Reshape((4, 64)))\n", " model.add(UpSampling1D(size=2))\n", " \n", " model.add(Conv1D(filters=64, kernel_size=3, strides=1, padding='same', \n", " dilation_rate=1, kernel_initializer='normal'))\n", " #model.add(BatchNormalization())\n", " model.add(Activation('relu'))\n", " #model.add(Dropout(0.2))\n", " #model.add(UpSampling1D(size=5))\n", "\n", " model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", " dilation_rate=1, kernel_initializer='normal'))\n", " model.add(Activation('relu'))\n", " #model.add(Dropout(0.3))\n", " \n", "# model.add(Conv1D(filters=64, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(Activation('relu')) \n", "# model.add(Dropout(0.3))\n", "\n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(Activation('relu')) \n", " \n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(Activation('relu')) \n", " \n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(Activation('relu')) \n", " \n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, kernel_initializer='normal'))\n", "# model.add(Activation('relu')) \n", " \n", " model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", " dilation_rate=1, kernel_initializer='normal'))\n", " model.add(Activation('relu')) \n", " \n", " \n", " \n", " model.add(Flatten())\n", " # Compile model\n", "# if num_gpus == 1:\n", " model.compile(loss=naive_percent_loss, optimizer='adam', metrics=[calc_mre_K])\n", "# else:\n", "# model.compile(loss=naive_percent_loss, optimizer='adam', metrics=['accuracy'], context = gpu_list)\n", " \n", " \n", " return model \n", "\n", "def resnetb():\n", " model = Sequential()\n", " \n", " #first layer6\n", " model.add(Dense(256, input_dim=8, kernel_initializer='normal', \n", " name='first'))\n", " model.add(PReLU(alpha_initializer='zeros', alpha_regularizer=None))\n", " model.add(Reshape((8, 32)))\n", " \n", " \n", " #resnet block\n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, \n", "# kernel_initializer='normal'))\n", "# model.add(PReLU(alpha_initializer='zeros', alpha_regularizer=None))\n", "# model.add(Conv1D(filters=32, kernel_size=3, strides=1, padding='same', \n", "# dilation_rate=1, \n", "# kernel_initializer='normal'))\n", " \n", " \n", " \n", " #Last layer\n", " model.add(Flatten())\n", " \n", " #compile model\n", " model.compile(loss=naive_percent_loss, optimizer='adam', metrics=[calc_mre_K])\n", " \n", " return model\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Model testing" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "ExecuteTime": { "end_time": "2018-10-10T10:25:37.434708Z", "start_time": "2018-10-10T08:57:24.329039Z" }, "code_folding": [], "scrolled": true }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/hegder/anaconda3/envs/dp2/lib/python3.5/site-packages/keras/backend/mxnet_backend.py:89: UserWarning: MXNet Backend performs best with `channels_first` format. Using `channels_last` will significantly reduce performance due to the Transpose operations. For performance improvement, please use this API`keras.utils.to_channels_first(x_input)`to transform `channels_last` data to `channels_first` format and also please change the `image_data_format` in `keras.json` to `channels_first`.Note: `x_input` is a Numpy tensor or a list of Numpy tensorRefer to: https://github.com/awslabs/keras-apache-mxnet/tree/master/docs/mxnet_backend/performance_guide.md\n", " train_symbol = func(*args, **kwargs)\n", "/home/hegder/anaconda3/envs/dp2/lib/python3.5/site-packages/keras/backend/mxnet_backend.py:92: UserWarning: MXNet Backend performs best with `channels_first` format. Using `channels_last` will significantly reduce performance due to the Transpose operations. For performance improvement, please use this API`keras.utils.to_channels_first(x_input)`to transform `channels_last` data to `channels_first` format and also please change the `image_data_format` in `keras.json` to `channels_first`.Note: `x_input` is a Numpy tensor or a list of Numpy tensorRefer to: https://github.com/awslabs/keras-apache-mxnet/tree/master/docs/mxnet_backend/performance_guide.md\n", " test_symbol = func(*args, **kwargs)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n", "Layer (type) Output Shape Param # \n", "=================================================================\n", "first (Dense) (None, 256) 2304 \n", "_________________________________________________________________\n", "p_re_lu_9 (PReLU) (None, 256) 256 \n", "_________________________________________________________________\n", "Reshape1 (Reshape) (None, 4, 64) 0 \n", "_________________________________________________________________\n", "Up1 (UpSampling1D) (None, 8, 64) 0 \n", "_________________________________________________________________\n", "Conv1 (Conv1D) (None, 8, 64) 12352 \n", "_________________________________________________________________\n", "p_re_lu_10 (PReLU) (None, 8, 64) 512 \n", "_________________________________________________________________\n", "Conv2 (Conv1D) (None, 8, 32) 6176 \n", "_________________________________________________________________\n", "p_re_lu_11 (PReLU) (None, 8, 32) 256 \n", "_________________________________________________________________\n", "Conv3 (Conv1D) (None, 8, 32) 3104 \n", "_________________________________________________________________\n", "p_re_lu_12 (PReLU) (None, 8, 32) 256 \n", "_________________________________________________________________\n", "Conv4 (Conv1D) (None, 8, 32) 3104 \n", "_________________________________________________________________\n", "p_re_lu_13 (PReLU) (None, 8, 32) 256 \n", "_________________________________________________________________\n", "Conv5 (Conv1D) (None, 8, 32) 3104 \n", "_________________________________________________________________\n", "p_re_lu_14 (PReLU) (None, 8, 32) 256 \n", "_________________________________________________________________\n", "flatten_6 (Flatten) (None, 256) 0 \n", "=================================================================\n", "Total params: 31,936\n", "Trainable params: 31,936\n", "Non-trainable params: 0\n", "_________________________________________________________________\n", "Train on 48000 samples, validate on 12000 samples\n", "Epoch 1/2000\n", " 3136/48000 [>.............................] - ETA: 3s - loss: 9075.2638 - calc_mre_K: 55.3910 " ] }, { "name": "stderr", "output_type": "stream", "text": [ "/home/hegder/anaconda3/envs/dp2/lib/python3.5/site-packages/mxnet/module/bucketing_module.py:408: UserWarning: Optimizer created manually outside Module but rescale_grad is not normalized to 1.0/batch_size/num_workers (1.0 vs. 0.015625). Is this intended?\n", " force_init=force_init)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 55us/step - loss: 2031.7968 - calc_mre_K: 12.4011 - val_loss: 846.2158 - val_calc_mre_K: 5.1718\n", "Epoch 2/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 654.2782 - calc_mre_K: 3.9934 - val_loss: 525.1828 - val_calc_mre_K: 3.2097\n", "Epoch 3/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 464.0548 - calc_mre_K: 2.8324 - val_loss: 431.2603 - val_calc_mre_K: 2.6357\n", "Epoch 4/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 415.3811 - calc_mre_K: 2.5353 - val_loss: 381.9067 - val_calc_mre_K: 2.3341\n", "Epoch 5/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 395.1123 - calc_mre_K: 2.4116 - val_loss: 423.0315 - val_calc_mre_K: 2.5854\n", "Epoch 6/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 361.1197 - calc_mre_K: 2.2041 - val_loss: 346.1212 - val_calc_mre_K: 2.1153\n", "Epoch 7/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 348.3622 - calc_mre_K: 2.1262 - val_loss: 386.5240 - val_calc_mre_K: 2.3624\n", "Epoch 8/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 341.7338 - calc_mre_K: 2.0858 - val_loss: 327.4623 - val_calc_mre_K: 2.0013\n", "Epoch 9/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 322.0901 - calc_mre_K: 1.9659 - val_loss: 335.6857 - val_calc_mre_K: 2.0516\n", "Epoch 10/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 311.4419 - calc_mre_K: 1.9009 - val_loss: 352.9701 - val_calc_mre_K: 2.1573\n", "Epoch 11/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 307.7178 - calc_mre_K: 1.8782 - val_loss: 276.0660 - val_calc_mre_K: 1.6872\n", "Epoch 12/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 297.4845 - calc_mre_K: 1.8157 - val_loss: 292.3918 - val_calc_mre_K: 1.7870\n", "Epoch 13/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 292.7277 - calc_mre_K: 1.7867 - val_loss: 265.1606 - val_calc_mre_K: 1.6206\n", "Epoch 14/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 283.2072 - calc_mre_K: 1.7286 - val_loss: 259.7420 - val_calc_mre_K: 1.5874\n", "Epoch 15/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 278.6771 - calc_mre_K: 1.7009 - val_loss: 258.4965 - val_calc_mre_K: 1.5799\n", "Epoch 16/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 270.6699 - calc_mre_K: 1.6520 - val_loss: 318.6926 - val_calc_mre_K: 1.9478\n", "Epoch 17/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 268.7820 - calc_mre_K: 1.6405 - val_loss: 292.4416 - val_calc_mre_K: 1.7873\n", "Epoch 18/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 260.7596 - calc_mre_K: 1.5916 - val_loss: 251.0446 - val_calc_mre_K: 1.5343\n", "Epoch 19/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 263.0421 - calc_mre_K: 1.6055 - val_loss: 305.9421 - val_calc_mre_K: 1.8698\n", "Epoch 20/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 253.6871 - calc_mre_K: 1.5484 - val_loss: 234.1713 - val_calc_mre_K: 1.4311\n", "Epoch 21/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 246.7939 - calc_mre_K: 1.5063 - val_loss: 228.7624 - val_calc_mre_K: 1.3981\n", "Epoch 22/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 243.4549 - calc_mre_K: 1.4859 - val_loss: 222.1738 - val_calc_mre_K: 1.3579\n", "Epoch 23/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 242.3187 - calc_mre_K: 1.4790 - val_loss: 298.7863 - val_calc_mre_K: 1.8260\n", "Epoch 24/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 235.6943 - calc_mre_K: 1.4386 - val_loss: 222.8858 - val_calc_mre_K: 1.3622\n", "Epoch 25/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 244.0877 - calc_mre_K: 1.4898 - val_loss: 209.9085 - val_calc_mre_K: 1.2830\n", "Epoch 26/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 227.0154 - calc_mre_K: 1.3856 - val_loss: 205.9902 - val_calc_mre_K: 1.2590\n", "Epoch 27/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 226.2994 - calc_mre_K: 1.3812 - val_loss: 213.9432 - val_calc_mre_K: 1.3076\n", "Epoch 28/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 227.0551 - calc_mre_K: 1.3858 - val_loss: 208.6655 - val_calc_mre_K: 1.2753\n", "Epoch 29/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 216.8113 - calc_mre_K: 1.3233 - val_loss: 205.1455 - val_calc_mre_K: 1.2538\n", "Epoch 30/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 219.2269 - calc_mre_K: 1.3381 - val_loss: 204.8767 - val_calc_mre_K: 1.2522\n", "Epoch 31/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 215.0880 - calc_mre_K: 1.3128 - val_loss: 201.7308 - val_calc_mre_K: 1.2329\n", "Epoch 32/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 219.0455 - calc_mre_K: 1.3369 - val_loss: 191.2438 - val_calc_mre_K: 1.1688\n", "Epoch 33/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 207.9515 - calc_mre_K: 1.2692 - val_loss: 187.8990 - val_calc_mre_K: 1.1484\n", "Epoch 34/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 213.4755 - calc_mre_K: 1.3030 - val_loss: 189.9531 - val_calc_mre_K: 1.1610\n", "Epoch 35/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 204.6645 - calc_mre_K: 1.2492 - val_loss: 183.7407 - val_calc_mre_K: 1.1230\n", "Epoch 36/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 197.7033 - calc_mre_K: 1.2067 - val_loss: 209.5619 - val_calc_mre_K: 1.2808\n", "Epoch 37/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 203.1189 - calc_mre_K: 1.2397 - val_loss: 298.9484 - val_calc_mre_K: 1.8271\n", "Epoch 38/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 201.4136 - calc_mre_K: 1.2293 - val_loss: 186.2745 - val_calc_mre_K: 1.1385\n", "Epoch 39/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 197.2231 - calc_mre_K: 1.2038 - val_loss: 209.0864 - val_calc_mre_K: 1.2779\n", "Epoch 40/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 196.6815 - calc_mre_K: 1.2004 - val_loss: 205.8515 - val_calc_mre_K: 1.2581\n", "Epoch 41/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 192.1387 - calc_mre_K: 1.1727 - val_loss: 193.3299 - val_calc_mre_K: 1.1816\n", "Epoch 42/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 192.3806 - calc_mre_K: 1.1742 - val_loss: 216.9004 - val_calc_mre_K: 1.3257\n", "Epoch 43/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 190.0321 - calc_mre_K: 1.1599 - val_loss: 211.8614 - val_calc_mre_K: 1.2949\n", "Epoch 44/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 189.6738 - calc_mre_K: 1.1577 - val_loss: 173.7575 - val_calc_mre_K: 1.0620\n", "Epoch 45/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 184.4621 - calc_mre_K: 1.1259 - val_loss: 209.7584 - val_calc_mre_K: 1.2821\n", "Epoch 46/2000\n", "48000/48000 [==============================] - 2s 49us/step - loss: 184.6972 - calc_mre_K: 1.1273 - val_loss: 173.5362 - val_calc_mre_K: 1.0606\n", "Epoch 47/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 181.4477 - calc_mre_K: 1.1075 - val_loss: 180.0106 - val_calc_mre_K: 1.1002\n", "Epoch 48/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 180.3946 - calc_mre_K: 1.1010 - val_loss: 180.7794 - val_calc_mre_K: 1.1049\n", "Epoch 49/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 179.7924 - calc_mre_K: 1.0974 - val_loss: 159.8966 - val_calc_mre_K: 0.9773\n", "Epoch 50/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 176.9709 - calc_mre_K: 1.0801 - val_loss: 189.2257 - val_calc_mre_K: 1.1565\n", "Epoch 51/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 176.0698 - calc_mre_K: 1.0746 - val_loss: 169.3423 - val_calc_mre_K: 1.0350\n", "Epoch 52/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 175.5652 - calc_mre_K: 1.0716 - val_loss: 155.3760 - val_calc_mre_K: 0.9496\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 53/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 173.2421 - calc_mre_K: 1.0574 - val_loss: 166.0553 - val_calc_mre_K: 1.0149\n", "Epoch 54/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 172.1844 - calc_mre_K: 1.0509 - val_loss: 159.2861 - val_calc_mre_K: 0.9735\n", "Epoch 55/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 171.0818 - calc_mre_K: 1.0442 - val_loss: 165.3388 - val_calc_mre_K: 1.0105\n", "Epoch 56/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 173.4710 - calc_mre_K: 1.0588 - val_loss: 225.7608 - val_calc_mre_K: 1.3798\n", "Epoch 57/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 164.7979 - calc_mre_K: 1.0058 - val_loss: 202.5197 - val_calc_mre_K: 1.2378\n", "Epoch 58/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 170.6269 - calc_mre_K: 1.0414 - val_loss: 168.4014 - val_calc_mre_K: 1.0292\n", "Epoch 59/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 164.1800 - calc_mre_K: 1.0021 - val_loss: 151.6793 - val_calc_mre_K: 0.9270\n", "Epoch 60/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 161.9741 - calc_mre_K: 0.9886 - val_loss: 186.0728 - val_calc_mre_K: 1.1372\n", "Epoch 61/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 161.3803 - calc_mre_K: 0.9850 - val_loss: 161.8322 - val_calc_mre_K: 0.9891\n", "Epoch 62/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 159.9662 - calc_mre_K: 0.9764 - val_loss: 161.7553 - val_calc_mre_K: 0.9886\n", "Epoch 63/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 159.3066 - calc_mre_K: 0.9723 - val_loss: 147.6567 - val_calc_mre_K: 0.9025\n", "Epoch 64/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 160.2363 - calc_mre_K: 0.9780 - val_loss: 156.7479 - val_calc_mre_K: 0.9580\n", "Epoch 65/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 156.5090 - calc_mre_K: 0.9553 - val_loss: 142.1342 - val_calc_mre_K: 0.8687\n", "Epoch 66/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 157.2742 - calc_mre_K: 0.9599 - val_loss: 144.0670 - val_calc_mre_K: 0.8805\n", "Epoch 67/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 154.1915 - calc_mre_K: 0.9411 - val_loss: 166.5235 - val_calc_mre_K: 1.0178\n", "Epoch 68/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 154.3540 - calc_mre_K: 0.9421 - val_loss: 163.8930 - val_calc_mre_K: 1.0017\n", "Epoch 69/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 152.7808 - calc_mre_K: 0.9325 - val_loss: 157.5231 - val_calc_mre_K: 0.9628\n", "Epoch 70/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 157.6668 - calc_mre_K: 0.9623 - val_loss: 168.6097 - val_calc_mre_K: 1.0305\n", "Epoch 71/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 149.4487 - calc_mre_K: 0.9122 - val_loss: 190.7350 - val_calc_mre_K: 1.1658\n", "Epoch 72/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 153.2022 - calc_mre_K: 0.9351 - val_loss: 176.5913 - val_calc_mre_K: 1.0793\n", "Epoch 73/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 149.1965 - calc_mre_K: 0.9106 - val_loss: 157.8075 - val_calc_mre_K: 0.9645\n", "Epoch 74/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 150.0736 - calc_mre_K: 0.9160 - val_loss: 145.1817 - val_calc_mre_K: 0.8873\n", "Epoch 75/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 149.6253 - calc_mre_K: 0.9132 - val_loss: 172.8706 - val_calc_mre_K: 1.0566\n", "Epoch 76/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 148.9916 - calc_mre_K: 0.9094 - val_loss: 152.1679 - val_calc_mre_K: 0.9301\n", "Epoch 77/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 147.0738 - calc_mre_K: 0.8977 - val_loss: 151.0286 - val_calc_mre_K: 0.9231\n", "Epoch 78/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 143.9739 - calc_mre_K: 0.8787 - val_loss: 134.7897 - val_calc_mre_K: 0.8238\n", "Epoch 79/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 148.2275 - calc_mre_K: 0.9047 - val_loss: 151.1734 - val_calc_mre_K: 0.9239\n", "Epoch 80/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 144.1350 - calc_mre_K: 0.8797 - val_loss: 144.1633 - val_calc_mre_K: 0.8811\n", "Epoch 81/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 142.9003 - calc_mre_K: 0.8722 - val_loss: 155.9298 - val_calc_mre_K: 0.9530\n", "Epoch 82/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 146.5374 - calc_mre_K: 0.8944 - val_loss: 138.1997 - val_calc_mre_K: 0.8446\n", "Epoch 83/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 145.1741 - calc_mre_K: 0.8861 - val_loss: 130.6899 - val_calc_mre_K: 0.7987\n", "Epoch 84/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 139.3203 - calc_mre_K: 0.8503 - val_loss: 128.3232 - val_calc_mre_K: 0.7843\n", "Epoch 85/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 142.4783 - calc_mre_K: 0.8696 - val_loss: 153.2325 - val_calc_mre_K: 0.9365\n", "Epoch 86/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 138.6833 - calc_mre_K: 0.8465 - val_loss: 183.7586 - val_calc_mre_K: 1.1231\n", "Epoch 87/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 138.9266 - calc_mre_K: 0.8479 - val_loss: 127.3387 - val_calc_mre_K: 0.7783\n", "Epoch 88/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 149.2663 - calc_mre_K: 0.9110 - val_loss: 130.2275 - val_calc_mre_K: 0.7959\n", "Epoch 89/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 134.8378 - calc_mre_K: 0.8230 - val_loss: 151.0780 - val_calc_mre_K: 0.9234\n", "Epoch 90/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 136.1563 - calc_mre_K: 0.8310 - val_loss: 141.0784 - val_calc_mre_K: 0.8623\n", "Epoch 91/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 137.3221 - calc_mre_K: 0.8381 - val_loss: 127.6931 - val_calc_mre_K: 0.7804\n", "Epoch 92/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 136.6010 - calc_mre_K: 0.8337 - val_loss: 151.5313 - val_calc_mre_K: 0.9261\n", "Epoch 93/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 136.8399 - calc_mre_K: 0.8352 - val_loss: 149.1418 - val_calc_mre_K: 0.9115\n", "Epoch 94/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 138.1458 - calc_mre_K: 0.8432 - val_loss: 133.9925 - val_calc_mre_K: 0.8189\n", "Epoch 95/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 133.7001 - calc_mre_K: 0.8160 - val_loss: 124.0457 - val_calc_mre_K: 0.7582\n", "Epoch 96/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 134.5207 - calc_mre_K: 0.8210 - val_loss: 144.0742 - val_calc_mre_K: 0.8806\n", "Epoch 97/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 134.4601 - calc_mre_K: 0.8207 - val_loss: 125.6598 - val_calc_mre_K: 0.7680\n", "Epoch 98/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 133.1603 - calc_mre_K: 0.8127 - val_loss: 140.0122 - val_calc_mre_K: 0.8558\n", "Epoch 99/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 140.2959 - calc_mre_K: 0.8563 - val_loss: 129.6088 - val_calc_mre_K: 0.7921\n", "Epoch 100/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 129.5315 - calc_mre_K: 0.7906 - val_loss: 135.9989 - val_calc_mre_K: 0.8312\n", "Epoch 101/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 132.9128 - calc_mre_K: 0.8112 - val_loss: 129.3815 - val_calc_mre_K: 0.7908\n", "Epoch 102/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 131.0593 - calc_mre_K: 0.7999 - val_loss: 126.5930 - val_calc_mre_K: 0.7737\n", "Epoch 103/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 129.9678 - calc_mre_K: 0.7933 - val_loss: 138.0317 - val_calc_mre_K: 0.8436\n", "Epoch 104/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 131.0791 - calc_mre_K: 0.8000 - val_loss: 128.0260 - val_calc_mre_K: 0.7825\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 105/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 129.4538 - calc_mre_K: 0.7901 - val_loss: 134.9206 - val_calc_mre_K: 0.8246\n", "Epoch 106/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 128.9063 - calc_mre_K: 0.7868 - val_loss: 126.2694 - val_calc_mre_K: 0.7717\n", "Epoch 107/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 132.4297 - calc_mre_K: 0.8083 - val_loss: 138.4491 - val_calc_mre_K: 0.8462\n", "Epoch 108/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 130.6471 - calc_mre_K: 0.7974 - val_loss: 124.5086 - val_calc_mre_K: 0.7610\n", "Epoch 109/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 127.2583 - calc_mre_K: 0.7767 - val_loss: 126.1941 - val_calc_mre_K: 0.7713\n", "Epoch 110/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 130.1550 - calc_mre_K: 0.7944 - val_loss: 131.2944 - val_calc_mre_K: 0.8025\n", "Epoch 111/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 127.9703 - calc_mre_K: 0.7811 - val_loss: 139.8529 - val_calc_mre_K: 0.8547\n", "Epoch 112/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 124.8063 - calc_mre_K: 0.7618 - val_loss: 126.0766 - val_calc_mre_K: 0.7706\n", "Epoch 113/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 125.9102 - calc_mre_K: 0.7685 - val_loss: 132.4511 - val_calc_mre_K: 0.8095\n", "Epoch 114/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 127.7734 - calc_mre_K: 0.7799 - val_loss: 141.1208 - val_calc_mre_K: 0.8625\n", "Epoch 115/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 126.1135 - calc_mre_K: 0.7697 - val_loss: 117.9223 - val_calc_mre_K: 0.7207\n", "Epoch 116/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 125.3857 - calc_mre_K: 0.7653 - val_loss: 116.0043 - val_calc_mre_K: 0.7090\n", "Epoch 117/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 125.4345 - calc_mre_K: 0.7656 - val_loss: 118.9414 - val_calc_mre_K: 0.7270\n", "Epoch 118/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 124.0140 - calc_mre_K: 0.7569 - val_loss: 116.1544 - val_calc_mre_K: 0.7099\n", "Epoch 119/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 126.4604 - calc_mre_K: 0.7719 - val_loss: 127.4505 - val_calc_mre_K: 0.7790\n", "Epoch 120/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 120.8862 - calc_mre_K: 0.7378 - val_loss: 117.0842 - val_calc_mre_K: 0.7156\n", "Epoch 121/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 123.1240 - calc_mre_K: 0.7515 - val_loss: 124.2454 - val_calc_mre_K: 0.7594\n", "Epoch 122/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 123.9544 - calc_mre_K: 0.7566 - val_loss: 120.6845 - val_calc_mre_K: 0.7376\n", "Epoch 123/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 123.3550 - calc_mre_K: 0.7529 - val_loss: 115.1120 - val_calc_mre_K: 0.7036\n", "Epoch 124/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 121.4785 - calc_mre_K: 0.7414 - val_loss: 126.1173 - val_calc_mre_K: 0.7708\n", "Epoch 125/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 120.0115 - calc_mre_K: 0.7325 - val_loss: 116.7531 - val_calc_mre_K: 0.7136\n", "Epoch 126/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 122.6092 - calc_mre_K: 0.7483 - val_loss: 121.7569 - val_calc_mre_K: 0.7442\n", "Epoch 127/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 121.3320 - calc_mre_K: 0.7406 - val_loss: 122.9598 - val_calc_mre_K: 0.7515\n", "Epoch 128/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 125.6976 - calc_mre_K: 0.7672 - val_loss: 112.5952 - val_calc_mre_K: 0.6882\n", "Epoch 129/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 117.1801 - calc_mre_K: 0.7152 - val_loss: 113.4377 - val_calc_mre_K: 0.6933\n", "Epoch 130/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 119.9985 - calc_mre_K: 0.7324 - val_loss: 121.3926 - val_calc_mre_K: 0.7419\n", "Epoch 131/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 119.7441 - calc_mre_K: 0.7309 - val_loss: 112.4827 - val_calc_mre_K: 0.6875\n", "Epoch 132/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 124.8075 - calc_mre_K: 0.7618 - val_loss: 136.3236 - val_calc_mre_K: 0.8332\n", "Epoch 133/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 119.1693 - calc_mre_K: 0.7274 - val_loss: 131.7057 - val_calc_mre_K: 0.8050\n", "Epoch 134/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 118.6338 - calc_mre_K: 0.7241 - val_loss: 115.1472 - val_calc_mre_K: 0.7038\n", "Epoch 135/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 120.5158 - calc_mre_K: 0.7356 - val_loss: 117.8278 - val_calc_mre_K: 0.7202\n", "Epoch 136/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 118.9856 - calc_mre_K: 0.7262 - val_loss: 117.0852 - val_calc_mre_K: 0.7156\n", "Epoch 137/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 118.7229 - calc_mre_K: 0.7246 - val_loss: 121.4629 - val_calc_mre_K: 0.7424\n", "Epoch 138/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 117.6352 - calc_mre_K: 0.7180 - val_loss: 142.6666 - val_calc_mre_K: 0.8719\n", "Epoch 139/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 117.4220 - calc_mre_K: 0.7167 - val_loss: 123.6894 - val_calc_mre_K: 0.7560\n", "Epoch 140/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 119.7145 - calc_mre_K: 0.7307 - val_loss: 121.6672 - val_calc_mre_K: 0.7436\n", "Epoch 141/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 119.4480 - calc_mre_K: 0.7291 - val_loss: 113.5554 - val_calc_mre_K: 0.6940\n", "Epoch 142/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 115.7756 - calc_mre_K: 0.7066 - val_loss: 114.5285 - val_calc_mre_K: 0.7000\n", "Epoch 143/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 123.6708 - calc_mre_K: 0.7548 - val_loss: 109.1084 - val_calc_mre_K: 0.6669\n", "Epoch 144/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 114.7168 - calc_mre_K: 0.7002 - val_loss: 118.4980 - val_calc_mre_K: 0.7242\n", "Epoch 145/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 116.0657 - calc_mre_K: 0.7084 - val_loss: 130.2837 - val_calc_mre_K: 0.7963\n", "Epoch 146/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 117.6108 - calc_mre_K: 0.7178 - val_loss: 110.9500 - val_calc_mre_K: 0.6781\n", "Epoch 147/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 117.7085 - calc_mre_K: 0.7184 - val_loss: 114.9590 - val_calc_mre_K: 0.7026\n", "Epoch 148/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 115.3832 - calc_mre_K: 0.7042 - val_loss: 129.1149 - val_calc_mre_K: 0.7891\n", "Epoch 149/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 114.4387 - calc_mre_K: 0.6985 - val_loss: 107.8438 - val_calc_mre_K: 0.6591\n", "Epoch 150/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 114.2158 - calc_mre_K: 0.6971 - val_loss: 110.5380 - val_calc_mre_K: 0.6756\n", "Epoch 151/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 114.7289 - calc_mre_K: 0.7002 - val_loss: 106.9668 - val_calc_mre_K: 0.6538\n", "Epoch 152/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 114.6049 - calc_mre_K: 0.6995 - val_loss: 109.2779 - val_calc_mre_K: 0.6679\n", "Epoch 153/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 115.9021 - calc_mre_K: 0.7074 - val_loss: 118.5315 - val_calc_mre_K: 0.7244\n", "Epoch 154/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 112.1789 - calc_mre_K: 0.6847 - val_loss: 110.0540 - val_calc_mre_K: 0.6726\n", "Epoch 155/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 113.4215 - calc_mre_K: 0.6923 - val_loss: 119.3028 - val_calc_mre_K: 0.7291\n", "Epoch 156/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 112.5986 - calc_mre_K: 0.6872 - val_loss: 111.1813 - val_calc_mre_K: 0.6795\n", "Epoch 157/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 115.0506 - calc_mre_K: 0.7022 - val_loss: 111.7231 - val_calc_mre_K: 0.6828\n", "Epoch 158/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 116.9169 - calc_mre_K: 0.7136 - val_loss: 117.5527 - val_calc_mre_K: 0.7185\n", "Epoch 159/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 112.1828 - calc_mre_K: 0.6847 - val_loss: 107.3782 - val_calc_mre_K: 0.6563\n", "Epoch 160/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 112.6083 - calc_mre_K: 0.6873 - val_loss: 155.9483 - val_calc_mre_K: 0.9531\n", "Epoch 161/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 116.9786 - calc_mre_K: 0.7140 - val_loss: 136.2801 - val_calc_mre_K: 0.8329\n", "Epoch 162/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 118.1842 - calc_mre_K: 0.7213 - val_loss: 109.4381 - val_calc_mre_K: 0.6689\n", "Epoch 163/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 108.3826 - calc_mre_K: 0.6615 - val_loss: 128.0515 - val_calc_mre_K: 0.7826\n", "Epoch 164/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 112.5101 - calc_mre_K: 0.6867 - val_loss: 107.9039 - val_calc_mre_K: 0.6595\n", "Epoch 165/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 111.6644 - calc_mre_K: 0.6815 - val_loss: 106.1721 - val_calc_mre_K: 0.6489\n", "Epoch 166/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 112.4412 - calc_mre_K: 0.6863 - val_loss: 113.4126 - val_calc_mre_K: 0.6932\n", "Epoch 167/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 111.3352 - calc_mre_K: 0.6795 - val_loss: 113.1599 - val_calc_mre_K: 0.6916\n", "Epoch 168/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 110.5161 - calc_mre_K: 0.6745 - val_loss: 106.4724 - val_calc_mre_K: 0.6508\n", "Epoch 169/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 115.0639 - calc_mre_K: 0.7023 - val_loss: 111.9164 - val_calc_mre_K: 0.6840\n", "Epoch 170/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 113.2308 - calc_mre_K: 0.6911 - val_loss: 104.3240 - val_calc_mre_K: 0.6376\n", "Epoch 171/2000\n", "48000/48000 [==============================] - 3s 63us/step - loss: 108.2361 - calc_mre_K: 0.6606 - val_loss: 106.5627 - val_calc_mre_K: 0.6513\n", "Epoch 172/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 111.3628 - calc_mre_K: 0.6797 - val_loss: 104.2172 - val_calc_mre_K: 0.6369\n", "Epoch 173/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 110.2008 - calc_mre_K: 0.6726 - val_loss: 120.7192 - val_calc_mre_K: 0.7378\n", "Epoch 174/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 109.2558 - calc_mre_K: 0.6668 - val_loss: 116.1290 - val_calc_mre_K: 0.7098\n", "Epoch 175/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 110.9368 - calc_mre_K: 0.6771 - val_loss: 115.6180 - val_calc_mre_K: 0.7066\n", "Epoch 176/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 111.5019 - calc_mre_K: 0.6806 - val_loss: 108.2289 - val_calc_mre_K: 0.6615\n", "Epoch 177/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 109.0701 - calc_mre_K: 0.6657 - val_loss: 118.9698 - val_calc_mre_K: 0.7271\n", "Epoch 178/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 111.2063 - calc_mre_K: 0.6787 - val_loss: 109.7707 - val_calc_mre_K: 0.6709\n", "Epoch 179/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 108.3974 - calc_mre_K: 0.6616 - val_loss: 121.2729 - val_calc_mre_K: 0.7412\n", "Epoch 180/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 109.0259 - calc_mre_K: 0.6654 - val_loss: 105.9601 - val_calc_mre_K: 0.6476\n", "Epoch 181/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 108.4829 - calc_mre_K: 0.6621 - val_loss: 111.1234 - val_calc_mre_K: 0.6792\n", "Epoch 182/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 108.8888 - calc_mre_K: 0.6646 - val_loss: 108.8840 - val_calc_mre_K: 0.6655\n", "Epoch 183/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 107.8335 - calc_mre_K: 0.6582 - val_loss: 104.4597 - val_calc_mre_K: 0.6385\n", "Epoch 184/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 108.9351 - calc_mre_K: 0.6649 - val_loss: 107.5189 - val_calc_mre_K: 0.6571\n", "Epoch 185/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 108.0461 - calc_mre_K: 0.6595 - val_loss: 107.5368 - val_calc_mre_K: 0.6573\n", "Epoch 186/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 109.8855 - calc_mre_K: 0.6707 - val_loss: 105.9254 - val_calc_mre_K: 0.6474\n", "Epoch 187/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 106.6187 - calc_mre_K: 0.6507 - val_loss: 107.8826 - val_calc_mre_K: 0.6594\n", "Epoch 188/2000\n", "48000/48000 [==============================] - 3s 64us/step - loss: 107.6476 - calc_mre_K: 0.6570 - val_loss: 101.6346 - val_calc_mre_K: 0.6212\n", "Epoch 189/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 108.1702 - calc_mre_K: 0.6602 - val_loss: 102.6145 - val_calc_mre_K: 0.6272\n", "Epoch 190/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 110.3900 - calc_mre_K: 0.6738 - val_loss: 100.1838 - val_calc_mre_K: 0.6123\n", "Epoch 191/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 108.7409 - calc_mre_K: 0.6637 - val_loss: 104.1784 - val_calc_mre_K: 0.6367\n", "Epoch 192/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 107.5158 - calc_mre_K: 0.6562 - val_loss: 99.0938 - val_calc_mre_K: 0.6056\n", "Epoch 193/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 105.2763 - calc_mre_K: 0.6426 - val_loss: 108.4849 - val_calc_mre_K: 0.6631\n", "Epoch 194/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 107.3527 - calc_mre_K: 0.6552 - val_loss: 104.8140 - val_calc_mre_K: 0.6406\n", "Epoch 195/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 106.4689 - calc_mre_K: 0.6498 - val_loss: 116.9961 - val_calc_mre_K: 0.7151\n", "Epoch 196/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 106.5508 - calc_mre_K: 0.6503 - val_loss: 103.1732 - val_calc_mre_K: 0.6306\n", "Epoch 197/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 106.6355 - calc_mre_K: 0.6509 - val_loss: 116.1842 - val_calc_mre_K: 0.7101\n", "Epoch 198/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 109.2703 - calc_mre_K: 0.6669 - val_loss: 137.1835 - val_calc_mre_K: 0.8384\n", "Epoch 199/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 104.5899 - calc_mre_K: 0.6384 - val_loss: 98.9977 - val_calc_mre_K: 0.6051\n", "Epoch 200/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 105.3478 - calc_mre_K: 0.6430 - val_loss: 97.9781 - val_calc_mre_K: 0.5988\n", "Epoch 201/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 106.7128 - calc_mre_K: 0.6513 - val_loss: 101.8863 - val_calc_mre_K: 0.6227\n", "Epoch 202/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 105.8224 - calc_mre_K: 0.6459 - val_loss: 113.5767 - val_calc_mre_K: 0.6942\n", "Epoch 203/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 105.3158 - calc_mre_K: 0.6428 - val_loss: 98.9878 - val_calc_mre_K: 0.6050\n", "Epoch 204/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 104.4533 - calc_mre_K: 0.6375 - val_loss: 104.8323 - val_calc_mre_K: 0.6407\n", "Epoch 205/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 104.9789 - calc_mre_K: 0.6407 - val_loss: 107.0195 - val_calc_mre_K: 0.6541\n", "Epoch 206/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 105.1520 - calc_mre_K: 0.6418 - val_loss: 126.6935 - val_calc_mre_K: 0.7743\n", "Epoch 207/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 54us/step - loss: 107.2713 - calc_mre_K: 0.6547 - val_loss: 105.9937 - val_calc_mre_K: 0.6478\n", "Epoch 208/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 104.1053 - calc_mre_K: 0.6354 - val_loss: 116.0131 - val_calc_mre_K: 0.7091\n", "Epoch 209/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 104.1980 - calc_mre_K: 0.6360 - val_loss: 113.6486 - val_calc_mre_K: 0.6946\n", "Epoch 210/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 104.6845 - calc_mre_K: 0.6389 - val_loss: 98.5809 - val_calc_mre_K: 0.6025\n", "Epoch 211/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 104.1774 - calc_mre_K: 0.6358 - val_loss: 98.9973 - val_calc_mre_K: 0.6051\n", "Epoch 212/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 103.9606 - calc_mre_K: 0.6345 - val_loss: 98.1034 - val_calc_mre_K: 0.5996\n", "Epoch 213/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 105.7821 - calc_mre_K: 0.6456 - val_loss: 107.6900 - val_calc_mre_K: 0.6582\n", "Epoch 214/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 103.3125 - calc_mre_K: 0.6306 - val_loss: 105.2994 - val_calc_mre_K: 0.6436\n", "Epoch 215/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 103.2225 - calc_mre_K: 0.6300 - val_loss: 106.5948 - val_calc_mre_K: 0.6515\n", "Epoch 216/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 103.2520 - calc_mre_K: 0.6302 - val_loss: 102.5168 - val_calc_mre_K: 0.6266\n", "Epoch 217/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 103.7100 - calc_mre_K: 0.6330 - val_loss: 99.3939 - val_calc_mre_K: 0.6075\n", "Epoch 218/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 103.8806 - calc_mre_K: 0.6340 - val_loss: 99.6751 - val_calc_mre_K: 0.6092\n", "Epoch 219/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 103.8708 - calc_mre_K: 0.6340 - val_loss: 99.2943 - val_calc_mre_K: 0.6069\n", "Epoch 220/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 102.7503 - calc_mre_K: 0.6271 - val_loss: 107.3268 - val_calc_mre_K: 0.6560\n", "Epoch 221/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 102.5865 - calc_mre_K: 0.6261 - val_loss: 103.0829 - val_calc_mre_K: 0.6300\n", "Epoch 222/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 104.7738 - calc_mre_K: 0.6395 - val_loss: 107.0252 - val_calc_mre_K: 0.6541\n", "Epoch 223/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 104.7632 - calc_mre_K: 0.6394 - val_loss: 101.2258 - val_calc_mre_K: 0.6187\n", "Epoch 224/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 102.8238 - calc_mre_K: 0.6276 - val_loss: 100.0737 - val_calc_mre_K: 0.6116\n", "Epoch 225/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 102.4481 - calc_mre_K: 0.6253 - val_loss: 105.0126 - val_calc_mre_K: 0.6418\n", "Epoch 226/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 102.7481 - calc_mre_K: 0.6271 - val_loss: 98.2394 - val_calc_mre_K: 0.6004\n", "Epoch 227/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 101.4092 - calc_mre_K: 0.6190 - val_loss: 100.4361 - val_calc_mre_K: 0.6138\n", "Epoch 228/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 101.9616 - calc_mre_K: 0.6223 - val_loss: 105.6130 - val_calc_mre_K: 0.6455\n", "Epoch 229/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 101.3986 - calc_mre_K: 0.6189 - val_loss: 97.0959 - val_calc_mre_K: 0.5934\n", "Epoch 230/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 102.8789 - calc_mre_K: 0.6279 - val_loss: 103.7793 - val_calc_mre_K: 0.6343\n", "Epoch 231/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 103.2092 - calc_mre_K: 0.6299 - val_loss: 96.8188 - val_calc_mre_K: 0.5918\n", "Epoch 232/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 100.2961 - calc_mre_K: 0.6122 - val_loss: 99.8270 - val_calc_mre_K: 0.6101\n", "Epoch 233/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 101.7756 - calc_mre_K: 0.6212 - val_loss: 97.8456 - val_calc_mre_K: 0.5980\n", "Epoch 234/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 99.9495 - calc_mre_K: 0.6100 - val_loss: 102.7109 - val_calc_mre_K: 0.6277\n", "Epoch 235/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 102.4736 - calc_mre_K: 0.6254 - val_loss: 113.4066 - val_calc_mre_K: 0.6931\n", "Epoch 236/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 101.2193 - calc_mre_K: 0.6178 - val_loss: 103.9540 - val_calc_mre_K: 0.6354\n", "Epoch 237/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 101.1878 - calc_mre_K: 0.6176 - val_loss: 97.2089 - val_calc_mre_K: 0.5941\n", "Epoch 238/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 100.3748 - calc_mre_K: 0.6126 - val_loss: 101.2232 - val_calc_mre_K: 0.6187\n", "Epoch 239/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 100.5513 - calc_mre_K: 0.6137 - val_loss: 103.5004 - val_calc_mre_K: 0.6326\n", "Epoch 240/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 101.6327 - calc_mre_K: 0.6203 - val_loss: 102.5386 - val_calc_mre_K: 0.6267\n", "Epoch 241/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 99.4312 - calc_mre_K: 0.6069 - val_loss: 108.1778 - val_calc_mre_K: 0.6612\n", "Epoch 242/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 102.6508 - calc_mre_K: 0.6265 - val_loss: 99.7813 - val_calc_mre_K: 0.6099\n", "Epoch 243/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 101.1440 - calc_mre_K: 0.6173 - val_loss: 105.8270 - val_calc_mre_K: 0.6468\n", "Epoch 244/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 100.1270 - calc_mre_K: 0.6111 - val_loss: 106.8452 - val_calc_mre_K: 0.6530\n", "Epoch 245/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 100.7068 - calc_mre_K: 0.6147 - val_loss: 105.3840 - val_calc_mre_K: 0.6441\n", "Epoch 246/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 101.0989 - calc_mre_K: 0.6171 - val_loss: 99.0358 - val_calc_mre_K: 0.6053\n", "Epoch 247/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 99.2124 - calc_mre_K: 0.6055 - val_loss: 111.6251 - val_calc_mre_K: 0.6822\n", "Epoch 248/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 100.2344 - calc_mre_K: 0.6118 - val_loss: 98.2925 - val_calc_mre_K: 0.6008\n", "Epoch 249/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 99.4008 - calc_mre_K: 0.6067 - val_loss: 104.0004 - val_calc_mre_K: 0.6356\n", "Epoch 250/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 100.3291 - calc_mre_K: 0.6124 - val_loss: 97.4749 - val_calc_mre_K: 0.5958\n", "Epoch 251/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 99.7145 - calc_mre_K: 0.6086 - val_loss: 102.6316 - val_calc_mre_K: 0.6273\n", "Epoch 252/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 101.1852 - calc_mre_K: 0.6176 - val_loss: 103.2534 - val_calc_mre_K: 0.6311\n", "Epoch 253/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 98.4727 - calc_mre_K: 0.6010 - val_loss: 100.7876 - val_calc_mre_K: 0.6160\n", "Epoch 254/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 97.7408 - calc_mre_K: 0.5966 - val_loss: 94.6522 - val_calc_mre_K: 0.5785\n", "Epoch 255/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 99.3823 - calc_mre_K: 0.6066 - val_loss: 112.2148 - val_calc_mre_K: 0.6858\n", "Epoch 256/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 98.9001 - calc_mre_K: 0.6036 - val_loss: 94.2538 - val_calc_mre_K: 0.5761\n", "Epoch 257/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 98.5463 - calc_mre_K: 0.6015 - val_loss: 109.8395 - val_calc_mre_K: 0.6713\n", "Epoch 258/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 103.1124 - calc_mre_K: 0.6293 - val_loss: 92.2682 - val_calc_mre_K: 0.5639\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 259/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 97.7796 - calc_mre_K: 0.5968 - val_loss: 95.2558 - val_calc_mre_K: 0.5822\n", "Epoch 260/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 98.3614 - calc_mre_K: 0.6004 - val_loss: 101.9461 - val_calc_mre_K: 0.6231\n", "Epoch 261/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 98.9988 - calc_mre_K: 0.6042 - val_loss: 93.9266 - val_calc_mre_K: 0.5741\n", "Epoch 262/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 98.5042 - calc_mre_K: 0.6012 - val_loss: 106.2772 - val_calc_mre_K: 0.6496\n", "Epoch 263/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 98.7353 - calc_mre_K: 0.6026 - val_loss: 93.3907 - val_calc_mre_K: 0.5708\n", "Epoch 264/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 99.5590 - calc_mre_K: 0.6077 - val_loss: 102.1510 - val_calc_mre_K: 0.6243\n", "Epoch 265/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 98.8988 - calc_mre_K: 0.6036 - val_loss: 129.4554 - val_calc_mre_K: 0.7912\n", "Epoch 266/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 99.5079 - calc_mre_K: 0.6073 - val_loss: 97.9943 - val_calc_mre_K: 0.5989\n", "Epoch 267/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 97.9737 - calc_mre_K: 0.5980 - val_loss: 111.8501 - val_calc_mre_K: 0.6836\n", "Epoch 268/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 97.6926 - calc_mre_K: 0.5963 - val_loss: 97.6146 - val_calc_mre_K: 0.5966\n", "Epoch 269/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 98.6807 - calc_mre_K: 0.6023 - val_loss: 94.8138 - val_calc_mre_K: 0.5795\n", "Epoch 270/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 97.0444 - calc_mre_K: 0.5923 - val_loss: 98.1128 - val_calc_mre_K: 0.5997\n", "Epoch 271/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 96.9939 - calc_mre_K: 0.5920 - val_loss: 93.6683 - val_calc_mre_K: 0.5725\n", "Epoch 272/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 97.8611 - calc_mre_K: 0.5973 - val_loss: 108.3316 - val_calc_mre_K: 0.6621\n", "Epoch 273/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 97.3368 - calc_mre_K: 0.5941 - val_loss: 103.5605 - val_calc_mre_K: 0.6330\n", "Epoch 274/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 98.6586 - calc_mre_K: 0.6022 - val_loss: 98.5691 - val_calc_mre_K: 0.6025\n", "Epoch 275/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 97.5807 - calc_mre_K: 0.5956 - val_loss: 93.3956 - val_calc_mre_K: 0.5708\n", "Epoch 276/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 97.1809 - calc_mre_K: 0.5931 - val_loss: 95.9035 - val_calc_mre_K: 0.5862\n", "Epoch 277/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 97.6068 - calc_mre_K: 0.5957 - val_loss: 133.3767 - val_calc_mre_K: 0.8152\n", "Epoch 278/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 97.5739 - calc_mre_K: 0.5955 - val_loss: 97.1606 - val_calc_mre_K: 0.5938\n", "Epoch 279/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 96.9104 - calc_mre_K: 0.5915 - val_loss: 95.2223 - val_calc_mre_K: 0.5820\n", "Epoch 280/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 96.7715 - calc_mre_K: 0.5906 - val_loss: 103.4723 - val_calc_mre_K: 0.6324\n", "Epoch 281/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 96.7170 - calc_mre_K: 0.5903 - val_loss: 102.9481 - val_calc_mre_K: 0.6292\n", "Epoch 282/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 97.2150 - calc_mre_K: 0.5934 - val_loss: 91.5260 - val_calc_mre_K: 0.5594\n", "Epoch 283/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 96.0122 - calc_mre_K: 0.5860 - val_loss: 92.3282 - val_calc_mre_K: 0.5643\n", "Epoch 284/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 97.8686 - calc_mre_K: 0.5973 - val_loss: 94.0045 - val_calc_mre_K: 0.5746\n", "Epoch 285/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 97.8025 - calc_mre_K: 0.5969 - val_loss: 96.1902 - val_calc_mre_K: 0.5879\n", "Epoch 286/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 95.3237 - calc_mre_K: 0.5818 - val_loss: 90.5050 - val_calc_mre_K: 0.5532\n", "Epoch 287/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 97.3628 - calc_mre_K: 0.5943 - val_loss: 116.4975 - val_calc_mre_K: 0.7120\n", "Epoch 288/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 95.7087 - calc_mre_K: 0.5842 - val_loss: 92.5928 - val_calc_mre_K: 0.5659\n", "Epoch 289/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 96.2468 - calc_mre_K: 0.5874 - val_loss: 91.5473 - val_calc_mre_K: 0.5595\n", "Epoch 290/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 96.5703 - calc_mre_K: 0.5894 - val_loss: 97.6954 - val_calc_mre_K: 0.5971\n", "Epoch 291/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 96.0027 - calc_mre_K: 0.5860 - val_loss: 93.9681 - val_calc_mre_K: 0.5743\n", "Epoch 292/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 96.5592 - calc_mre_K: 0.5894 - val_loss: 91.2492 - val_calc_mre_K: 0.5577\n", "Epoch 293/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 96.0834 - calc_mre_K: 0.5864 - val_loss: 101.1353 - val_calc_mre_K: 0.6182\n", "Epoch 294/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 95.0193 - calc_mre_K: 0.5800 - val_loss: 95.2718 - val_calc_mre_K: 0.5823\n", "Epoch 295/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 96.8362 - calc_mre_K: 0.5910 - val_loss: 93.5318 - val_calc_mre_K: 0.5717\n", "Epoch 296/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 94.9524 - calc_mre_K: 0.5795 - val_loss: 98.1200 - val_calc_mre_K: 0.5997\n", "Epoch 297/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 96.4588 - calc_mre_K: 0.5887 - val_loss: 103.5516 - val_calc_mre_K: 0.6329\n", "Epoch 298/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 95.0082 - calc_mre_K: 0.5799 - val_loss: 101.1237 - val_calc_mre_K: 0.6181\n", "Epoch 299/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 95.8574 - calc_mre_K: 0.5851 - val_loss: 101.2952 - val_calc_mre_K: 0.6191\n", "Epoch 300/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 96.0622 - calc_mre_K: 0.5863 - val_loss: 92.7715 - val_calc_mre_K: 0.5670\n", "Epoch 301/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 94.7637 - calc_mre_K: 0.5784 - val_loss: 101.3908 - val_calc_mre_K: 0.6197\n", "Epoch 302/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 96.7859 - calc_mre_K: 0.5907 - val_loss: 98.8808 - val_calc_mre_K: 0.6043\n", "Epoch 303/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 94.9678 - calc_mre_K: 0.5796 - val_loss: 101.5788 - val_calc_mre_K: 0.6209\n", "Epoch 304/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 95.0182 - calc_mre_K: 0.5799 - val_loss: 102.0076 - val_calc_mre_K: 0.6235\n", "Epoch 305/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 94.1595 - calc_mre_K: 0.5747 - val_loss: 96.2856 - val_calc_mre_K: 0.5885\n", "Epoch 306/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 95.5843 - calc_mre_K: 0.5834 - val_loss: 102.9839 - val_calc_mre_K: 0.6294\n", "Epoch 307/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 94.8920 - calc_mre_K: 0.5792 - val_loss: 103.9699 - val_calc_mre_K: 0.6354\n", "Epoch 308/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 96.0936 - calc_mre_K: 0.5865 - val_loss: 92.2093 - val_calc_mre_K: 0.5636\n", "Epoch 309/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 93.6362 - calc_mre_K: 0.5715 - val_loss: 91.4590 - val_calc_mre_K: 0.5590\n", "Epoch 310/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 95.4239 - calc_mre_K: 0.5824 - val_loss: 94.5206 - val_calc_mre_K: 0.5777\n", "Epoch 311/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 58us/step - loss: 94.7157 - calc_mre_K: 0.5781 - val_loss: 97.2685 - val_calc_mre_K: 0.5945\n", "Epoch 312/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 94.9409 - calc_mre_K: 0.5795 - val_loss: 92.7366 - val_calc_mre_K: 0.5668\n", "Epoch 313/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 95.1319 - calc_mre_K: 0.5806 - val_loss: 97.1883 - val_calc_mre_K: 0.5940\n", "Epoch 314/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 94.1500 - calc_mre_K: 0.5746 - val_loss: 91.8530 - val_calc_mre_K: 0.5614\n", "Epoch 315/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 95.3901 - calc_mre_K: 0.5822 - val_loss: 90.1071 - val_calc_mre_K: 0.5508\n", "Epoch 316/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 93.4605 - calc_mre_K: 0.5704 - val_loss: 89.9605 - val_calc_mre_K: 0.5498\n", "Epoch 317/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 94.9125 - calc_mre_K: 0.5793 - val_loss: 92.1942 - val_calc_mre_K: 0.5635\n", "Epoch 318/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 94.2107 - calc_mre_K: 0.5750 - val_loss: 91.2939 - val_calc_mre_K: 0.5580\n", "Epoch 319/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 93.1959 - calc_mre_K: 0.5688 - val_loss: 100.9155 - val_calc_mre_K: 0.6168\n", "Epoch 320/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 94.3613 - calc_mre_K: 0.5759 - val_loss: 117.8747 - val_calc_mre_K: 0.7204\n", "Epoch 321/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 93.8500 - calc_mre_K: 0.5728 - val_loss: 92.4418 - val_calc_mre_K: 0.5650\n", "Epoch 322/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 93.2509 - calc_mre_K: 0.5692 - val_loss: 89.8440 - val_calc_mre_K: 0.5491\n", "Epoch 323/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 93.9385 - calc_mre_K: 0.5734 - val_loss: 94.8516 - val_calc_mre_K: 0.5797\n", "Epoch 324/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 92.9107 - calc_mre_K: 0.5671 - val_loss: 91.0978 - val_calc_mre_K: 0.5568\n", "Epoch 325/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 93.3327 - calc_mre_K: 0.5697 - val_loss: 89.2175 - val_calc_mre_K: 0.5453\n", "Epoch 326/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 93.0871 - calc_mre_K: 0.5682 - val_loss: 105.2763 - val_calc_mre_K: 0.6435\n", "Epoch 327/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 92.5557 - calc_mre_K: 0.5649 - val_loss: 94.2732 - val_calc_mre_K: 0.5762\n", "Epoch 328/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 93.3335 - calc_mre_K: 0.5697 - val_loss: 94.2632 - val_calc_mre_K: 0.5761\n", "Epoch 329/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 92.5188 - calc_mre_K: 0.5647 - val_loss: 88.4345 - val_calc_mre_K: 0.5405\n", "Epoch 330/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 92.7189 - calc_mre_K: 0.5659 - val_loss: 97.0592 - val_calc_mre_K: 0.5932\n", "Epoch 331/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 92.2102 - calc_mre_K: 0.5628 - val_loss: 92.7367 - val_calc_mre_K: 0.5668\n", "Epoch 332/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 92.6301 - calc_mre_K: 0.5654 - val_loss: 100.1732 - val_calc_mre_K: 0.6123\n", "Epoch 333/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 93.0042 - calc_mre_K: 0.5677 - val_loss: 91.6137 - val_calc_mre_K: 0.5599\n", "Epoch 334/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 92.7066 - calc_mre_K: 0.5658 - val_loss: 88.3972 - val_calc_mre_K: 0.5403\n", "Epoch 335/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 91.4549 - calc_mre_K: 0.5582 - val_loss: 94.3870 - val_calc_mre_K: 0.5769\n", "Epoch 336/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 92.5356 - calc_mre_K: 0.5648 - val_loss: 89.6609 - val_calc_mre_K: 0.5480\n", "Epoch 337/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 93.2076 - calc_mre_K: 0.5689 - val_loss: 90.1991 - val_calc_mre_K: 0.5513\n", "Epoch 338/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 92.7503 - calc_mre_K: 0.5661 - val_loss: 93.6879 - val_calc_mre_K: 0.5726\n", "Epoch 339/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 91.1826 - calc_mre_K: 0.5565 - val_loss: 89.8868 - val_calc_mre_K: 0.5494\n", "Epoch 340/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 92.3266 - calc_mre_K: 0.5635 - val_loss: 101.3557 - val_calc_mre_K: 0.6195\n", "Epoch 341/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 91.0978 - calc_mre_K: 0.5560 - val_loss: 90.7229 - val_calc_mre_K: 0.5545\n", "Epoch 342/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 92.8033 - calc_mre_K: 0.5664 - val_loss: 86.5991 - val_calc_mre_K: 0.5293\n", "Epoch 343/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 91.0989 - calc_mre_K: 0.5560 - val_loss: 89.9293 - val_calc_mre_K: 0.5497\n", "Epoch 344/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 91.7350 - calc_mre_K: 0.5599 - val_loss: 95.7807 - val_calc_mre_K: 0.5854\n", "Epoch 345/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 91.8777 - calc_mre_K: 0.5608 - val_loss: 90.0325 - val_calc_mre_K: 0.5503\n", "Epoch 346/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 91.2318 - calc_mre_K: 0.5568 - val_loss: 88.8347 - val_calc_mre_K: 0.5430\n", "Epoch 347/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 91.5170 - calc_mre_K: 0.5586 - val_loss: 87.1975 - val_calc_mre_K: 0.5330\n", "Epoch 348/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 91.5662 - calc_mre_K: 0.5589 - val_loss: 94.6583 - val_calc_mre_K: 0.5786\n", "Epoch 349/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 90.4416 - calc_mre_K: 0.5520 - val_loss: 87.6799 - val_calc_mre_K: 0.5359\n", "Epoch 350/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 90.1631 - calc_mre_K: 0.5503 - val_loss: 93.8964 - val_calc_mre_K: 0.5739\n", "Epoch 351/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 91.2465 - calc_mre_K: 0.5569 - val_loss: 101.4253 - val_calc_mre_K: 0.6199\n", "Epoch 352/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 90.7091 - calc_mre_K: 0.5536 - val_loss: 87.7623 - val_calc_mre_K: 0.5364\n", "Epoch 353/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 90.6012 - calc_mre_K: 0.5530 - val_loss: 88.4711 - val_calc_mre_K: 0.5407\n", "Epoch 354/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 90.2842 - calc_mre_K: 0.5511 - val_loss: 86.4502 - val_calc_mre_K: 0.5284\n", "Epoch 355/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 90.5705 - calc_mre_K: 0.5528 - val_loss: 98.0920 - val_calc_mre_K: 0.5995\n", "Epoch 356/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 90.6138 - calc_mre_K: 0.5531 - val_loss: 84.7595 - val_calc_mre_K: 0.5181\n", "Epoch 357/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 90.1874 - calc_mre_K: 0.5505 - val_loss: 90.5330 - val_calc_mre_K: 0.5533\n", "Epoch 358/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 90.2410 - calc_mre_K: 0.5508 - val_loss: 96.1584 - val_calc_mre_K: 0.5877\n", "Epoch 359/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 90.2556 - calc_mre_K: 0.5509 - val_loss: 91.5600 - val_calc_mre_K: 0.5596\n", "Epoch 360/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 90.7204 - calc_mre_K: 0.5537 - val_loss: 87.0255 - val_calc_mre_K: 0.5319\n", "Epoch 361/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 90.2336 - calc_mre_K: 0.5507 - val_loss: 95.1810 - val_calc_mre_K: 0.5818\n", "Epoch 362/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 88.7747 - calc_mre_K: 0.5418 - val_loss: 86.7078 - val_calc_mre_K: 0.5300\n", "Epoch 363/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 89.7833 - calc_mre_K: 0.5480 - val_loss: 89.0812 - val_calc_mre_K: 0.5445\n", "Epoch 364/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 89.3670 - calc_mre_K: 0.5455 - val_loss: 94.0466 - val_calc_mre_K: 0.5748\n", "Epoch 365/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 90.1886 - calc_mre_K: 0.5505 - val_loss: 88.6687 - val_calc_mre_K: 0.5420\n", "Epoch 366/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 89.5927 - calc_mre_K: 0.5468 - val_loss: 85.7970 - val_calc_mre_K: 0.5244\n", "Epoch 367/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 89.4138 - calc_mre_K: 0.5457 - val_loss: 88.3251 - val_calc_mre_K: 0.5398\n", "Epoch 368/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 89.7773 - calc_mre_K: 0.5480 - val_loss: 101.2593 - val_calc_mre_K: 0.6189\n", "Epoch 369/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 89.4140 - calc_mre_K: 0.5457 - val_loss: 85.4042 - val_calc_mre_K: 0.5220\n", "Epoch 370/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 89.8781 - calc_mre_K: 0.5486 - val_loss: 92.1434 - val_calc_mre_K: 0.5632\n", "Epoch 371/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 90.0802 - calc_mre_K: 0.5498 - val_loss: 92.1600 - val_calc_mre_K: 0.5633\n", "Epoch 372/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 87.7952 - calc_mre_K: 0.5359 - val_loss: 84.0283 - val_calc_mre_K: 0.5136\n", "Epoch 373/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 90.4799 - calc_mre_K: 0.5522 - val_loss: 92.7128 - val_calc_mre_K: 0.5667\n", "Epoch 374/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 88.8368 - calc_mre_K: 0.5422 - val_loss: 84.5482 - val_calc_mre_K: 0.5168\n", "Epoch 375/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 89.7015 - calc_mre_K: 0.5475 - val_loss: 87.6585 - val_calc_mre_K: 0.5358\n", "Epoch 376/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 87.9653 - calc_mre_K: 0.5369 - val_loss: 90.2071 - val_calc_mre_K: 0.5514\n", "Epoch 377/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 88.7150 - calc_mre_K: 0.5415 - val_loss: 87.6337 - val_calc_mre_K: 0.5356\n", "Epoch 378/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 89.0747 - calc_mre_K: 0.5437 - val_loss: 87.1346 - val_calc_mre_K: 0.5326\n", "Epoch 379/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 88.4073 - calc_mre_K: 0.5396 - val_loss: 91.1397 - val_calc_mre_K: 0.5571\n", "Epoch 380/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 88.6852 - calc_mre_K: 0.5413 - val_loss: 88.6249 - val_calc_mre_K: 0.5417\n", "Epoch 381/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 88.6530 - calc_mre_K: 0.5411 - val_loss: 84.9553 - val_calc_mre_K: 0.5192\n", "Epoch 382/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 89.2410 - calc_mre_K: 0.5447 - val_loss: 91.0755 - val_calc_mre_K: 0.5566\n", "Epoch 383/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 88.0598 - calc_mre_K: 0.5375 - val_loss: 84.4238 - val_calc_mre_K: 0.5160\n", "Epoch 384/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 90.1449 - calc_mre_K: 0.5502 - val_loss: 87.3620 - val_calc_mre_K: 0.5340\n", "Epoch 385/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 88.2701 - calc_mre_K: 0.5388 - val_loss: 96.8147 - val_calc_mre_K: 0.5917\n", "Epoch 386/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 88.1564 - calc_mre_K: 0.5381 - val_loss: 88.9063 - val_calc_mre_K: 0.5434\n", "Epoch 387/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 88.0993 - calc_mre_K: 0.5377 - val_loss: 86.6567 - val_calc_mre_K: 0.5296\n", "Epoch 388/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 87.2289 - calc_mre_K: 0.5324 - val_loss: 96.8141 - val_calc_mre_K: 0.5917\n", "Epoch 389/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 87.7328 - calc_mre_K: 0.5355 - val_loss: 85.4228 - val_calc_mre_K: 0.5221\n", "Epoch 390/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 87.5567 - calc_mre_K: 0.5344 - val_loss: 83.8738 - val_calc_mre_K: 0.5126\n", "Epoch 391/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 88.4356 - calc_mre_K: 0.5398 - val_loss: 97.4720 - val_calc_mre_K: 0.5957\n", "Epoch 392/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 87.2792 - calc_mre_K: 0.5327 - val_loss: 85.9580 - val_calc_mre_K: 0.5254\n", "Epoch 393/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 87.8419 - calc_mre_K: 0.5361 - val_loss: 88.5067 - val_calc_mre_K: 0.5409\n", "Epoch 394/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 88.3274 - calc_mre_K: 0.5391 - val_loss: 84.3061 - val_calc_mre_K: 0.5153\n", "Epoch 395/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 86.2035 - calc_mre_K: 0.5261 - val_loss: 85.3485 - val_calc_mre_K: 0.5216\n", "Epoch 396/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 88.2279 - calc_mre_K: 0.5385 - val_loss: 107.8934 - val_calc_mre_K: 0.6594\n", "Epoch 397/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 88.2784 - calc_mre_K: 0.5388 - val_loss: 83.7369 - val_calc_mre_K: 0.5118\n", "Epoch 398/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 86.7303 - calc_mre_K: 0.5294 - val_loss: 87.1277 - val_calc_mre_K: 0.5325\n", "Epoch 399/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 88.2834 - calc_mre_K: 0.5388 - val_loss: 84.1289 - val_calc_mre_K: 0.5142\n", "Epoch 400/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 87.3980 - calc_mre_K: 0.5334 - val_loss: 83.7419 - val_calc_mre_K: 0.5118\n", "Epoch 401/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 86.9538 - calc_mre_K: 0.5307 - val_loss: 90.6205 - val_calc_mre_K: 0.5539\n", "Epoch 402/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 87.2380 - calc_mre_K: 0.5325 - val_loss: 116.4075 - val_calc_mre_K: 0.7115\n", "Epoch 403/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 87.0975 - calc_mre_K: 0.5316 - val_loss: 82.9935 - val_calc_mre_K: 0.5073\n", "Epoch 404/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 87.1362 - calc_mre_K: 0.5318 - val_loss: 90.5322 - val_calc_mre_K: 0.5533\n", "Epoch 405/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 86.9243 - calc_mre_K: 0.5305 - val_loss: 89.8038 - val_calc_mre_K: 0.5489\n", "Epoch 406/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 86.5836 - calc_mre_K: 0.5285 - val_loss: 86.6624 - val_calc_mre_K: 0.5297\n", "Epoch 407/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 86.8295 - calc_mre_K: 0.5300 - val_loss: 86.5516 - val_calc_mre_K: 0.5290\n", "Epoch 408/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 86.6075 - calc_mre_K: 0.5286 - val_loss: 83.9382 - val_calc_mre_K: 0.5130\n", "Epoch 409/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 87.0606 - calc_mre_K: 0.5314 - val_loss: 86.0920 - val_calc_mre_K: 0.5262\n", "Epoch 410/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 87.4829 - calc_mre_K: 0.5340 - val_loss: 83.4511 - val_calc_mre_K: 0.5100\n", "Epoch 411/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 85.4411 - calc_mre_K: 0.5215 - val_loss: 81.1049 - val_calc_mre_K: 0.4957\n", "Epoch 412/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 87.8864 - calc_mre_K: 0.5364 - val_loss: 88.7739 - val_calc_mre_K: 0.5426\n", "Epoch 413/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 86.1666 - calc_mre_K: 0.5259 - val_loss: 98.6226 - val_calc_mre_K: 0.6028\n", "Epoch 414/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 86.4469 - calc_mre_K: 0.5276 - val_loss: 89.3229 - val_calc_mre_K: 0.5459\n", "Epoch 415/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 55us/step - loss: 86.1281 - calc_mre_K: 0.5257 - val_loss: 81.7868 - val_calc_mre_K: 0.4999\n", "Epoch 416/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 87.3378 - calc_mre_K: 0.5331 - val_loss: 87.1821 - val_calc_mre_K: 0.5329\n", "Epoch 417/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 85.7697 - calc_mre_K: 0.5235 - val_loss: 84.2322 - val_calc_mre_K: 0.5148\n", "Epoch 418/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 86.9440 - calc_mre_K: 0.5307 - val_loss: 80.7191 - val_calc_mre_K: 0.4934\n", "Epoch 419/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 85.7107 - calc_mre_K: 0.5231 - val_loss: 91.9072 - val_calc_mre_K: 0.5617\n", "Epoch 420/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 87.1697 - calc_mre_K: 0.5320 - val_loss: 83.1460 - val_calc_mre_K: 0.5082\n", "Epoch 421/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 85.4485 - calc_mre_K: 0.5215 - val_loss: 97.0125 - val_calc_mre_K: 0.5929\n", "Epoch 422/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 86.1508 - calc_mre_K: 0.5258 - val_loss: 89.4279 - val_calc_mre_K: 0.5466\n", "Epoch 423/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 85.6378 - calc_mre_K: 0.5227 - val_loss: 84.1954 - val_calc_mre_K: 0.5146\n", "Epoch 424/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 87.8616 - calc_mre_K: 0.5363 - val_loss: 82.2074 - val_calc_mre_K: 0.5025\n", "Epoch 425/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 86.0017 - calc_mre_K: 0.5249 - val_loss: 84.9558 - val_calc_mre_K: 0.5192\n", "Epoch 426/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 85.3053 - calc_mre_K: 0.5207 - val_loss: 89.8768 - val_calc_mre_K: 0.5493\n", "Epoch 427/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 86.3437 - calc_mre_K: 0.5270 - val_loss: 83.1792 - val_calc_mre_K: 0.5084\n", "Epoch 428/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 86.1976 - calc_mre_K: 0.5261 - val_loss: 86.8104 - val_calc_mre_K: 0.5306\n", "Epoch 429/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 85.7302 - calc_mre_K: 0.5233 - val_loss: 83.1264 - val_calc_mre_K: 0.5081\n", "Epoch 430/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 85.8995 - calc_mre_K: 0.5243 - val_loss: 87.2585 - val_calc_mre_K: 0.5333\n", "Epoch 431/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 85.5814 - calc_mre_K: 0.5223 - val_loss: 84.9903 - val_calc_mre_K: 0.5195\n", "Epoch 432/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 84.5603 - calc_mre_K: 0.5161 - val_loss: 100.3753 - val_calc_mre_K: 0.6135\n", "Epoch 433/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 85.5442 - calc_mre_K: 0.5221 - val_loss: 85.1709 - val_calc_mre_K: 0.5206\n", "Epoch 434/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 87.3665 - calc_mre_K: 0.5332 - val_loss: 81.0826 - val_calc_mre_K: 0.4956\n", "Epoch 435/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 84.5459 - calc_mre_K: 0.5160 - val_loss: 88.5179 - val_calc_mre_K: 0.5410\n", "Epoch 436/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 84.6390 - calc_mre_K: 0.5166 - val_loss: 86.3177 - val_calc_mre_K: 0.5276\n", "Epoch 437/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 86.1771 - calc_mre_K: 0.5260 - val_loss: 85.0312 - val_calc_mre_K: 0.5197\n", "Epoch 438/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 86.5730 - calc_mre_K: 0.5284 - val_loss: 83.5725 - val_calc_mre_K: 0.5108\n", "Epoch 439/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 84.4953 - calc_mre_K: 0.5157 - val_loss: 91.7935 - val_calc_mre_K: 0.5610\n", "Epoch 440/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 86.0143 - calc_mre_K: 0.5250 - val_loss: 82.6815 - val_calc_mre_K: 0.5053\n", "Epoch 441/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 85.9021 - calc_mre_K: 0.5243 - val_loss: 80.1743 - val_calc_mre_K: 0.4900\n", "Epoch 442/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 85.2081 - calc_mre_K: 0.5201 - val_loss: 84.4680 - val_calc_mre_K: 0.5163\n", "Epoch 443/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 85.0739 - calc_mre_K: 0.5192 - val_loss: 84.7360 - val_calc_mre_K: 0.5179\n", "Epoch 444/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 86.1451 - calc_mre_K: 0.5258 - val_loss: 81.3928 - val_calc_mre_K: 0.4975\n", "Epoch 445/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 85.1125 - calc_mre_K: 0.5195 - val_loss: 85.2016 - val_calc_mre_K: 0.5208\n", "Epoch 446/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 85.4369 - calc_mre_K: 0.5215 - val_loss: 92.8650 - val_calc_mre_K: 0.5676\n", "Epoch 447/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 85.3252 - calc_mre_K: 0.5208 - val_loss: 84.1305 - val_calc_mre_K: 0.5142\n", "Epoch 448/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 84.7476 - calc_mre_K: 0.5173 - val_loss: 84.8277 - val_calc_mre_K: 0.5185\n", "Epoch 449/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 85.0708 - calc_mre_K: 0.5192 - val_loss: 83.6723 - val_calc_mre_K: 0.5114\n", "Epoch 450/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 86.6296 - calc_mre_K: 0.5287 - val_loss: 84.6506 - val_calc_mre_K: 0.5174\n", "Epoch 451/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 84.5529 - calc_mre_K: 0.5161 - val_loss: 85.1800 - val_calc_mre_K: 0.5206\n", "Epoch 452/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 83.9318 - calc_mre_K: 0.5123 - val_loss: 81.8940 - val_calc_mre_K: 0.5005\n", "Epoch 453/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 87.8645 - calc_mre_K: 0.5363 - val_loss: 88.8637 - val_calc_mre_K: 0.5431\n", "Epoch 454/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 84.0896 - calc_mre_K: 0.5132 - val_loss: 82.4184 - val_calc_mre_K: 0.5037\n", "Epoch 455/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 85.1558 - calc_mre_K: 0.5197 - val_loss: 82.0331 - val_calc_mre_K: 0.5014\n", "Epoch 456/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 83.9807 - calc_mre_K: 0.5126 - val_loss: 83.3873 - val_calc_mre_K: 0.5097\n", "Epoch 457/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 86.0693 - calc_mre_K: 0.5253 - val_loss: 87.9198 - val_calc_mre_K: 0.5374\n", "Epoch 458/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.7595 - calc_mre_K: 0.5112 - val_loss: 85.5376 - val_calc_mre_K: 0.5228\n", "Epoch 459/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 84.6983 - calc_mre_K: 0.5170 - val_loss: 91.6371 - val_calc_mre_K: 0.5601\n", "Epoch 460/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 84.7082 - calc_mre_K: 0.5170 - val_loss: 83.2270 - val_calc_mre_K: 0.5087\n", "Epoch 461/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 85.5920 - calc_mre_K: 0.5224 - val_loss: 85.3176 - val_calc_mre_K: 0.5215\n", "Epoch 462/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 84.4486 - calc_mre_K: 0.5154 - val_loss: 86.4508 - val_calc_mre_K: 0.5284\n", "Epoch 463/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.5030 - calc_mre_K: 0.5097 - val_loss: 87.4991 - val_calc_mre_K: 0.5348\n", "Epoch 464/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 85.2919 - calc_mre_K: 0.5206 - val_loss: 93.5097 - val_calc_mre_K: 0.5715\n", "Epoch 465/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.8637 - calc_mre_K: 0.5119 - val_loss: 81.7735 - val_calc_mre_K: 0.4998\n", "Epoch 466/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 85.4420 - calc_mre_K: 0.5215 - val_loss: 81.6360 - val_calc_mre_K: 0.4990\n", "Epoch 467/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.2775 - calc_mre_K: 0.5083 - val_loss: 85.2872 - val_calc_mre_K: 0.5213\n", "Epoch 468/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.9919 - calc_mre_K: 0.5126 - val_loss: 83.0853 - val_calc_mre_K: 0.5078\n", "Epoch 469/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 84.3268 - calc_mre_K: 0.5147 - val_loss: 85.1279 - val_calc_mre_K: 0.5203\n", "Epoch 470/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 84.2186 - calc_mre_K: 0.5140 - val_loss: 82.2682 - val_calc_mre_K: 0.5028\n", "Epoch 471/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 84.0503 - calc_mre_K: 0.5130 - val_loss: 80.6153 - val_calc_mre_K: 0.4927\n", "Epoch 472/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 84.2782 - calc_mre_K: 0.5144 - val_loss: 83.2106 - val_calc_mre_K: 0.5086\n", "Epoch 473/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.7608 - calc_mre_K: 0.5112 - val_loss: 96.6040 - val_calc_mre_K: 0.5904\n", "Epoch 474/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 84.3560 - calc_mre_K: 0.5149 - val_loss: 86.3066 - val_calc_mre_K: 0.5275\n", "Epoch 475/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 83.7136 - calc_mre_K: 0.5109 - val_loss: 81.6127 - val_calc_mre_K: 0.4988\n", "Epoch 476/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 84.2249 - calc_mre_K: 0.5141 - val_loss: 87.7590 - val_calc_mre_K: 0.5364\n", "Epoch 477/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.7920 - calc_mre_K: 0.5114 - val_loss: 84.2578 - val_calc_mre_K: 0.5150\n", "Epoch 478/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.6933 - calc_mre_K: 0.5108 - val_loss: 81.5756 - val_calc_mre_K: 0.4986\n", "Epoch 479/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.2581 - calc_mre_K: 0.5082 - val_loss: 79.9499 - val_calc_mre_K: 0.4886\n", "Epoch 480/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 85.5454 - calc_mre_K: 0.5221 - val_loss: 84.3220 - val_calc_mre_K: 0.5154\n", "Epoch 481/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.3718 - calc_mre_K: 0.5089 - val_loss: 87.5075 - val_calc_mre_K: 0.5348\n", "Epoch 482/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.9169 - calc_mre_K: 0.5061 - val_loss: 86.1260 - val_calc_mre_K: 0.5264\n", "Epoch 483/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.5237 - calc_mre_K: 0.5098 - val_loss: 86.5096 - val_calc_mre_K: 0.5288\n", "Epoch 484/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 83.7505 - calc_mre_K: 0.5112 - val_loss: 82.4629 - val_calc_mre_K: 0.5040\n", "Epoch 485/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 84.0362 - calc_mre_K: 0.5129 - val_loss: 92.8840 - val_calc_mre_K: 0.5677\n", "Epoch 486/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.0563 - calc_mre_K: 0.5069 - val_loss: 82.6111 - val_calc_mre_K: 0.5049\n", "Epoch 487/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 84.1965 - calc_mre_K: 0.5139 - val_loss: 79.0961 - val_calc_mre_K: 0.4834\n", "Epoch 488/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 85.3863 - calc_mre_K: 0.5212 - val_loss: 89.8554 - val_calc_mre_K: 0.5492\n", "Epoch 489/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.1476 - calc_mre_K: 0.5075 - val_loss: 79.3151 - val_calc_mre_K: 0.4848\n", "Epoch 490/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 82.9155 - calc_mre_K: 0.5061 - val_loss: 85.8200 - val_calc_mre_K: 0.5245\n", "Epoch 491/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 82.6227 - calc_mre_K: 0.5043 - val_loss: 81.1502 - val_calc_mre_K: 0.4960\n", "Epoch 492/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.9593 - calc_mre_K: 0.5124 - val_loss: 85.1021 - val_calc_mre_K: 0.5201\n", "Epoch 493/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.3567 - calc_mre_K: 0.5088 - val_loss: 85.2754 - val_calc_mre_K: 0.5212\n", "Epoch 494/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.2103 - calc_mre_K: 0.5079 - val_loss: 83.4257 - val_calc_mre_K: 0.5099\n", "Epoch 495/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.8083 - calc_mre_K: 0.5115 - val_loss: 90.8006 - val_calc_mre_K: 0.5550\n", "Epoch 496/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 83.1967 - calc_mre_K: 0.5078 - val_loss: 86.4662 - val_calc_mre_K: 0.5285\n", "Epoch 497/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.4602 - calc_mre_K: 0.5094 - val_loss: 83.0749 - val_calc_mre_K: 0.5078\n", "Epoch 498/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.6090 - calc_mre_K: 0.5103 - val_loss: 79.8715 - val_calc_mre_K: 0.4882\n", "Epoch 499/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 82.3954 - calc_mre_K: 0.5029 - val_loss: 80.1800 - val_calc_mre_K: 0.4901\n", "Epoch 500/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 83.2327 - calc_mre_K: 0.5080 - val_loss: 80.4743 - val_calc_mre_K: 0.4919\n", "Epoch 501/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 82.9366 - calc_mre_K: 0.5062 - val_loss: 78.9504 - val_calc_mre_K: 0.4825\n", "Epoch 502/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 82.3516 - calc_mre_K: 0.5026 - val_loss: 80.2665 - val_calc_mre_K: 0.4906\n", "Epoch 503/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.0073 - calc_mre_K: 0.5066 - val_loss: 81.3618 - val_calc_mre_K: 0.4973\n", "Epoch 504/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.5460 - calc_mre_K: 0.5099 - val_loss: 84.3630 - val_calc_mre_K: 0.5156\n", "Epoch 505/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 83.6091 - calc_mre_K: 0.5103 - val_loss: 84.9917 - val_calc_mre_K: 0.5195\n", "Epoch 506/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 82.3652 - calc_mre_K: 0.5027 - val_loss: 84.1671 - val_calc_mre_K: 0.5144\n", "Epoch 507/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 83.2323 - calc_mre_K: 0.5080 - val_loss: 85.2114 - val_calc_mre_K: 0.5208\n", "Epoch 508/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.3020 - calc_mre_K: 0.5084 - val_loss: 80.9458 - val_calc_mre_K: 0.4947\n", "Epoch 509/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 83.0967 - calc_mre_K: 0.5072 - val_loss: 87.6718 - val_calc_mre_K: 0.5358\n", "Epoch 510/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 82.3635 - calc_mre_K: 0.5027 - val_loss: 83.0081 - val_calc_mre_K: 0.5073\n", "Epoch 511/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 83.9002 - calc_mre_K: 0.5121 - val_loss: 80.5956 - val_calc_mre_K: 0.4926\n", "Epoch 512/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 83.2251 - calc_mre_K: 0.5080 - val_loss: 80.6882 - val_calc_mre_K: 0.4931\n", "Epoch 513/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.5800 - calc_mre_K: 0.5040 - val_loss: 79.3737 - val_calc_mre_K: 0.4851\n", "Epoch 514/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 81.7302 - calc_mre_K: 0.4988 - val_loss: 84.5862 - val_calc_mre_K: 0.5170\n", "Epoch 515/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 81.3463 - calc_mre_K: 0.4965 - val_loss: 81.7151 - val_calc_mre_K: 0.4994\n", "Epoch 516/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 83.6800 - calc_mre_K: 0.5107 - val_loss: 80.9045 - val_calc_mre_K: 0.4945\n", "Epoch 517/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.7222 - calc_mre_K: 0.5049 - val_loss: 84.7237 - val_calc_mre_K: 0.5178\n", "Epoch 518/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 82.9277 - calc_mre_K: 0.5062 - val_loss: 86.0216 - val_calc_mre_K: 0.5258\n", "Epoch 519/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 54us/step - loss: 82.1966 - calc_mre_K: 0.5017 - val_loss: 83.0163 - val_calc_mre_K: 0.5074\n", "Epoch 520/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 83.3114 - calc_mre_K: 0.5085 - val_loss: 85.7307 - val_calc_mre_K: 0.5240\n", "Epoch 521/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 81.4254 - calc_mre_K: 0.4970 - val_loss: 78.6196 - val_calc_mre_K: 0.4805\n", "Epoch 522/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 82.6848 - calc_mre_K: 0.5047 - val_loss: 79.9891 - val_calc_mre_K: 0.4889\n", "Epoch 523/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 85.8060 - calc_mre_K: 0.5237 - val_loss: 83.4820 - val_calc_mre_K: 0.5102\n", "Epoch 524/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 81.4702 - calc_mre_K: 0.4973 - val_loss: 84.0198 - val_calc_mre_K: 0.5135\n", "Epoch 525/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 81.4444 - calc_mre_K: 0.4971 - val_loss: 81.4524 - val_calc_mre_K: 0.4978\n", "Epoch 526/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 82.4379 - calc_mre_K: 0.5032 - val_loss: 84.5943 - val_calc_mre_K: 0.5170\n", "Epoch 527/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.3692 - calc_mre_K: 0.5027 - val_loss: 79.8655 - val_calc_mre_K: 0.4881\n", "Epoch 528/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.1266 - calc_mre_K: 0.5013 - val_loss: 78.5305 - val_calc_mre_K: 0.4800\n", "Epoch 529/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.5660 - calc_mre_K: 0.5039 - val_loss: 79.3839 - val_calc_mre_K: 0.4852\n", "Epoch 530/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.6178 - calc_mre_K: 0.5043 - val_loss: 80.1505 - val_calc_mre_K: 0.4899\n", "Epoch 531/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 81.8131 - calc_mre_K: 0.4993 - val_loss: 86.5171 - val_calc_mre_K: 0.5288\n", "Epoch 532/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 82.4098 - calc_mre_K: 0.5030 - val_loss: 83.0202 - val_calc_mre_K: 0.5074\n", "Epoch 533/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 82.3002 - calc_mre_K: 0.5023 - val_loss: 78.3309 - val_calc_mre_K: 0.4788\n", "Epoch 534/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 84.0218 - calc_mre_K: 0.5128 - val_loss: 79.2494 - val_calc_mre_K: 0.4844\n", "Epoch 535/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.0070 - calc_mre_K: 0.5005 - val_loss: 82.8439 - val_calc_mre_K: 0.5063\n", "Epoch 536/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.8092 - calc_mre_K: 0.4993 - val_loss: 80.2009 - val_calc_mre_K: 0.4902\n", "Epoch 537/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 81.8590 - calc_mre_K: 0.4996 - val_loss: 81.2032 - val_calc_mre_K: 0.4963\n", "Epoch 538/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.9889 - calc_mre_K: 0.5004 - val_loss: 87.9155 - val_calc_mre_K: 0.5374\n", "Epoch 539/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 81.9139 - calc_mre_K: 0.5000 - val_loss: 82.3821 - val_calc_mre_K: 0.5035\n", "Epoch 540/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 82.6502 - calc_mre_K: 0.5045 - val_loss: 82.0116 - val_calc_mre_K: 0.5012\n", "Epoch 541/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 82.1952 - calc_mre_K: 0.5017 - val_loss: 81.1104 - val_calc_mre_K: 0.4957\n", "Epoch 542/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 81.7613 - calc_mre_K: 0.4990 - val_loss: 79.4436 - val_calc_mre_K: 0.4856\n", "Epoch 543/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 81.3065 - calc_mre_K: 0.4963 - val_loss: 85.0340 - val_calc_mre_K: 0.5197\n", "Epoch 544/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 82.7846 - calc_mre_K: 0.5053 - val_loss: 78.9411 - val_calc_mre_K: 0.4825\n", "Epoch 545/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.7680 - calc_mre_K: 0.4991 - val_loss: 83.1211 - val_calc_mre_K: 0.5080\n", "Epoch 546/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.3712 - calc_mre_K: 0.4967 - val_loss: 82.3009 - val_calc_mre_K: 0.5030\n", "Epoch 547/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 83.3816 - calc_mre_K: 0.5089 - val_loss: 84.5878 - val_calc_mre_K: 0.5170\n", "Epoch 548/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 80.9637 - calc_mre_K: 0.4942 - val_loss: 84.3940 - val_calc_mre_K: 0.5158\n", "Epoch 549/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.0966 - calc_mre_K: 0.5011 - val_loss: 86.7935 - val_calc_mre_K: 0.5305\n", "Epoch 550/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.3550 - calc_mre_K: 0.4966 - val_loss: 77.7594 - val_calc_mre_K: 0.4753\n", "Epoch 551/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.7732 - calc_mre_K: 0.4930 - val_loss: 78.8973 - val_calc_mre_K: 0.4822\n", "Epoch 552/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 81.8376 - calc_mre_K: 0.4995 - val_loss: 82.6864 - val_calc_mre_K: 0.5054\n", "Epoch 553/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.9743 - calc_mre_K: 0.5003 - val_loss: 85.1407 - val_calc_mre_K: 0.5204\n", "Epoch 554/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 81.6313 - calc_mre_K: 0.4982 - val_loss: 78.4022 - val_calc_mre_K: 0.4792\n", "Epoch 555/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 82.1144 - calc_mre_K: 0.5012 - val_loss: 79.7151 - val_calc_mre_K: 0.4872\n", "Epoch 556/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.5409 - calc_mre_K: 0.4977 - val_loss: 87.4226 - val_calc_mre_K: 0.5343\n", "Epoch 557/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 82.1519 - calc_mre_K: 0.5014 - val_loss: 81.9643 - val_calc_mre_K: 0.5009\n", "Epoch 558/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.2810 - calc_mre_K: 0.4961 - val_loss: 85.5646 - val_calc_mre_K: 0.5230\n", "Epoch 559/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 81.1619 - calc_mre_K: 0.4954 - val_loss: 76.4447 - val_calc_mre_K: 0.4672\n", "Epoch 560/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.6965 - calc_mre_K: 0.4986 - val_loss: 81.6381 - val_calc_mre_K: 0.4990\n", "Epoch 561/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.4993 - calc_mre_K: 0.4974 - val_loss: 81.6246 - val_calc_mre_K: 0.4989\n", "Epoch 562/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.1850 - calc_mre_K: 0.4955 - val_loss: 85.2887 - val_calc_mre_K: 0.5213\n", "Epoch 563/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 81.9885 - calc_mre_K: 0.5004 - val_loss: 82.2912 - val_calc_mre_K: 0.5030\n", "Epoch 564/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.0537 - calc_mre_K: 0.4947 - val_loss: 82.0331 - val_calc_mre_K: 0.5014\n", "Epoch 565/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.2100 - calc_mre_K: 0.4957 - val_loss: 89.7412 - val_calc_mre_K: 0.5485\n", "Epoch 566/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.7583 - calc_mre_K: 0.4990 - val_loss: 86.5181 - val_calc_mre_K: 0.5288\n", "Epoch 567/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 82.0772 - calc_mre_K: 0.5010 - val_loss: 81.0483 - val_calc_mre_K: 0.4954\n", "Epoch 568/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 82.6173 - calc_mre_K: 0.5043 - val_loss: 78.4625 - val_calc_mre_K: 0.4796\n", "Epoch 569/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 80.8394 - calc_mre_K: 0.4934 - val_loss: 80.8846 - val_calc_mre_K: 0.4944\n", "Epoch 570/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.5891 - calc_mre_K: 0.4980 - val_loss: 77.4761 - val_calc_mre_K: 0.4735\n", "Epoch 571/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 81.3178 - calc_mre_K: 0.4963 - val_loss: 79.6932 - val_calc_mre_K: 0.4871\n", "Epoch 572/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 81.3808 - calc_mre_K: 0.4967 - val_loss: 84.2491 - val_calc_mre_K: 0.5149\n", "Epoch 573/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.8657 - calc_mre_K: 0.4936 - val_loss: 79.3225 - val_calc_mre_K: 0.4848\n", "Epoch 574/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.0303 - calc_mre_K: 0.4946 - val_loss: 80.7052 - val_calc_mre_K: 0.4933\n", "Epoch 575/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 81.5832 - calc_mre_K: 0.4979 - val_loss: 74.5917 - val_calc_mre_K: 0.4559\n", "Epoch 576/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.6877 - calc_mre_K: 0.4925 - val_loss: 83.6111 - val_calc_mre_K: 0.5110\n", "Epoch 577/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.3834 - calc_mre_K: 0.4967 - val_loss: 78.9105 - val_calc_mre_K: 0.4823\n", "Epoch 578/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.2387 - calc_mre_K: 0.4958 - val_loss: 78.3372 - val_calc_mre_K: 0.4788\n", "Epoch 579/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 80.8287 - calc_mre_K: 0.4933 - val_loss: 82.1588 - val_calc_mre_K: 0.5021\n", "Epoch 580/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 82.4633 - calc_mre_K: 0.5033 - val_loss: 80.4068 - val_calc_mre_K: 0.4914\n", "Epoch 581/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 80.3534 - calc_mre_K: 0.4904 - val_loss: 93.9048 - val_calc_mre_K: 0.5739\n", "Epoch 582/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 81.3732 - calc_mre_K: 0.4967 - val_loss: 79.5917 - val_calc_mre_K: 0.4865\n", "Epoch 583/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.0572 - calc_mre_K: 0.4947 - val_loss: 82.9862 - val_calc_mre_K: 0.5072\n", "Epoch 584/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 81.5071 - calc_mre_K: 0.4975 - val_loss: 76.7395 - val_calc_mre_K: 0.4690\n", "Epoch 585/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 80.4533 - calc_mre_K: 0.4910 - val_loss: 81.9444 - val_calc_mre_K: 0.5009\n", "Epoch 586/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.4733 - calc_mre_K: 0.4973 - val_loss: 84.4408 - val_calc_mre_K: 0.5161\n", "Epoch 587/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.4088 - calc_mre_K: 0.4908 - val_loss: 76.5830 - val_calc_mre_K: 0.4681\n", "Epoch 588/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 82.1292 - calc_mre_K: 0.5013 - val_loss: 82.6964 - val_calc_mre_K: 0.5054\n", "Epoch 589/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 80.3195 - calc_mre_K: 0.4902 - val_loss: 83.0805 - val_calc_mre_K: 0.5078\n", "Epoch 590/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 81.2370 - calc_mre_K: 0.4958 - val_loss: 83.7073 - val_calc_mre_K: 0.5116\n", "Epoch 591/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.1948 - calc_mre_K: 0.4895 - val_loss: 84.5261 - val_calc_mre_K: 0.5166\n", "Epoch 592/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 81.0818 - calc_mre_K: 0.4949 - val_loss: 79.1382 - val_calc_mre_K: 0.4837\n", "Epoch 593/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 80.3901 - calc_mre_K: 0.4907 - val_loss: 82.0347 - val_calc_mre_K: 0.5014\n", "Epoch 594/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 80.7756 - calc_mre_K: 0.4930 - val_loss: 79.7610 - val_calc_mre_K: 0.4875\n", "Epoch 595/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.4685 - calc_mre_K: 0.4911 - val_loss: 76.3259 - val_calc_mre_K: 0.4665\n", "Epoch 596/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.5833 - calc_mre_K: 0.4918 - val_loss: 81.1470 - val_calc_mre_K: 0.4960\n", "Epoch 597/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.5185 - calc_mre_K: 0.4914 - val_loss: 78.4995 - val_calc_mre_K: 0.4798\n", "Epoch 598/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 81.1250 - calc_mre_K: 0.4951 - val_loss: 81.2592 - val_calc_mre_K: 0.4966\n", "Epoch 599/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.3335 - calc_mre_K: 0.4903 - val_loss: 80.9331 - val_calc_mre_K: 0.4947\n", "Epoch 600/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.7937 - calc_mre_K: 0.4931 - val_loss: 79.8868 - val_calc_mre_K: 0.4883\n", "Epoch 601/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 81.1831 - calc_mre_K: 0.4955 - val_loss: 76.1072 - val_calc_mre_K: 0.4652\n", "Epoch 602/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.9989 - calc_mre_K: 0.4883 - val_loss: 78.3541 - val_calc_mre_K: 0.4789\n", "Epoch 603/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 80.1762 - calc_mre_K: 0.4894 - val_loss: 79.5704 - val_calc_mre_K: 0.4863\n", "Epoch 604/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 81.1273 - calc_mre_K: 0.4952 - val_loss: 79.0936 - val_calc_mre_K: 0.4834\n", "Epoch 605/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.0146 - calc_mre_K: 0.4884 - val_loss: 78.7459 - val_calc_mre_K: 0.4813\n", "Epoch 606/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 80.1781 - calc_mre_K: 0.4894 - val_loss: 77.7397 - val_calc_mre_K: 0.4751\n", "Epoch 607/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.1149 - calc_mre_K: 0.4890 - val_loss: 84.3336 - val_calc_mre_K: 0.5154\n", "Epoch 608/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 80.3935 - calc_mre_K: 0.4907 - val_loss: 78.3934 - val_calc_mre_K: 0.4791\n", "Epoch 609/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 80.8851 - calc_mre_K: 0.4937 - val_loss: 76.8955 - val_calc_mre_K: 0.4700\n", "Epoch 610/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.2385 - calc_mre_K: 0.4897 - val_loss: 79.6368 - val_calc_mre_K: 0.4867\n", "Epoch 611/2000\n", "48000/48000 [==============================] - 3s 63us/step - loss: 79.8892 - calc_mre_K: 0.4876 - val_loss: 87.4161 - val_calc_mre_K: 0.5343\n", "Epoch 612/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 80.0011 - calc_mre_K: 0.4883 - val_loss: 83.4761 - val_calc_mre_K: 0.5102\n", "Epoch 613/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 80.6588 - calc_mre_K: 0.4923 - val_loss: 87.0933 - val_calc_mre_K: 0.5323\n", "Epoch 614/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.9346 - calc_mre_K: 0.4940 - val_loss: 76.0511 - val_calc_mre_K: 0.4648\n", "Epoch 615/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.5584 - calc_mre_K: 0.4917 - val_loss: 84.3133 - val_calc_mre_K: 0.5153\n", "Epoch 616/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.4814 - calc_mre_K: 0.4912 - val_loss: 86.1900 - val_calc_mre_K: 0.5268\n", "Epoch 617/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.5464 - calc_mre_K: 0.4916 - val_loss: 80.9391 - val_calc_mre_K: 0.4947\n", "Epoch 618/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 80.3514 - calc_mre_K: 0.4904 - val_loss: 78.7335 - val_calc_mre_K: 0.4812\n", "Epoch 619/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.8439 - calc_mre_K: 0.4873 - val_loss: 80.1272 - val_calc_mre_K: 0.4897\n", "Epoch 620/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 80.4705 - calc_mre_K: 0.4912 - val_loss: 90.3706 - val_calc_mre_K: 0.5524\n", "Epoch 621/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 80.8473 - calc_mre_K: 0.4935 - val_loss: 76.3072 - val_calc_mre_K: 0.4664\n", "Epoch 622/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 79.9253 - calc_mre_K: 0.4878 - val_loss: 77.1704 - val_calc_mre_K: 0.4717\n", "Epoch 623/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 55us/step - loss: 80.0478 - calc_mre_K: 0.4886 - val_loss: 79.0295 - val_calc_mre_K: 0.4830\n", "Epoch 624/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.2394 - calc_mre_K: 0.4836 - val_loss: 79.8674 - val_calc_mre_K: 0.4882\n", "Epoch 625/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 80.7100 - calc_mre_K: 0.4926 - val_loss: 75.9482 - val_calc_mre_K: 0.4642\n", "Epoch 626/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.7827 - calc_mre_K: 0.4931 - val_loss: 81.3494 - val_calc_mre_K: 0.4972\n", "Epoch 627/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.2760 - calc_mre_K: 0.4839 - val_loss: 79.6280 - val_calc_mre_K: 0.4867\n", "Epoch 628/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 80.1940 - calc_mre_K: 0.4895 - val_loss: 80.0756 - val_calc_mre_K: 0.4894\n", "Epoch 629/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 78.8412 - calc_mre_K: 0.4812 - val_loss: 76.3114 - val_calc_mre_K: 0.4664\n", "Epoch 630/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 80.2321 - calc_mre_K: 0.4897 - val_loss: 80.8802 - val_calc_mre_K: 0.4943\n", "Epoch 631/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.0518 - calc_mre_K: 0.4886 - val_loss: 81.6671 - val_calc_mre_K: 0.4991\n", "Epoch 632/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.4725 - calc_mre_K: 0.4912 - val_loss: 90.0203 - val_calc_mre_K: 0.5502\n", "Epoch 633/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.6867 - calc_mre_K: 0.4864 - val_loss: 79.9215 - val_calc_mre_K: 0.4885\n", "Epoch 634/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 80.2553 - calc_mre_K: 0.4898 - val_loss: 78.1728 - val_calc_mre_K: 0.4778\n", "Epoch 635/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 79.5887 - calc_mre_K: 0.4858 - val_loss: 84.4680 - val_calc_mre_K: 0.5163\n", "Epoch 636/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 79.4706 - calc_mre_K: 0.4850 - val_loss: 77.6921 - val_calc_mre_K: 0.4748\n", "Epoch 637/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.5529 - calc_mre_K: 0.4917 - val_loss: 76.7210 - val_calc_mre_K: 0.4689\n", "Epoch 638/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 79.6079 - calc_mre_K: 0.4859 - val_loss: 79.7641 - val_calc_mre_K: 0.4875\n", "Epoch 639/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.3914 - calc_mre_K: 0.4907 - val_loss: 81.1951 - val_calc_mre_K: 0.4963\n", "Epoch 640/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 79.6163 - calc_mre_K: 0.4859 - val_loss: 75.9763 - val_calc_mre_K: 0.4644\n", "Epoch 641/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 80.6730 - calc_mre_K: 0.4924 - val_loss: 78.4504 - val_calc_mre_K: 0.4795\n", "Epoch 642/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 79.1667 - calc_mre_K: 0.4832 - val_loss: 75.9270 - val_calc_mre_K: 0.4641\n", "Epoch 643/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 78.9610 - calc_mre_K: 0.4819 - val_loss: 76.8529 - val_calc_mre_K: 0.4697\n", "Epoch 644/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.9553 - calc_mre_K: 0.4819 - val_loss: 76.2848 - val_calc_mre_K: 0.4662\n", "Epoch 645/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 79.7120 - calc_mre_K: 0.4865 - val_loss: 80.5576 - val_calc_mre_K: 0.4924\n", "Epoch 646/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.4258 - calc_mre_K: 0.4848 - val_loss: 77.7310 - val_calc_mre_K: 0.4751\n", "Epoch 647/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 80.1866 - calc_mre_K: 0.4894 - val_loss: 78.0930 - val_calc_mre_K: 0.4773\n", "Epoch 648/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 79.5865 - calc_mre_K: 0.4858 - val_loss: 76.5358 - val_calc_mre_K: 0.4678\n", "Epoch 649/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 79.0697 - calc_mre_K: 0.4826 - val_loss: 84.2162 - val_calc_mre_K: 0.5147\n", "Epoch 650/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 79.0024 - calc_mre_K: 0.4822 - val_loss: 80.3655 - val_calc_mre_K: 0.4912\n", "Epoch 651/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 79.7815 - calc_mre_K: 0.4869 - val_loss: 79.7566 - val_calc_mre_K: 0.4875\n", "Epoch 652/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.4595 - calc_mre_K: 0.4850 - val_loss: 80.5442 - val_calc_mre_K: 0.4923\n", "Epoch 653/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 79.4533 - calc_mre_K: 0.4849 - val_loss: 75.8171 - val_calc_mre_K: 0.4634\n", "Epoch 654/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 79.8417 - calc_mre_K: 0.4873 - val_loss: 76.0952 - val_calc_mre_K: 0.4651\n", "Epoch 655/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 79.0672 - calc_mre_K: 0.4826 - val_loss: 87.0572 - val_calc_mre_K: 0.5321\n", "Epoch 656/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 80.0095 - calc_mre_K: 0.4883 - val_loss: 74.3479 - val_calc_mre_K: 0.4544\n", "Epoch 657/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 79.4160 - calc_mre_K: 0.4847 - val_loss: 74.7215 - val_calc_mre_K: 0.4567\n", "Epoch 658/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.2127 - calc_mre_K: 0.4835 - val_loss: 77.5612 - val_calc_mre_K: 0.4740\n", "Epoch 659/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.4374 - calc_mre_K: 0.4848 - val_loss: 81.7346 - val_calc_mre_K: 0.4996\n", "Epoch 660/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 79.6887 - calc_mre_K: 0.4864 - val_loss: 76.7382 - val_calc_mre_K: 0.4690\n", "Epoch 661/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 81.5048 - calc_mre_K: 0.4975 - val_loss: 76.9656 - val_calc_mre_K: 0.4704\n", "Epoch 662/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 78.8454 - calc_mre_K: 0.4812 - val_loss: 76.2471 - val_calc_mre_K: 0.4660\n", "Epoch 663/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 78.9642 - calc_mre_K: 0.4820 - val_loss: 78.4202 - val_calc_mre_K: 0.4793\n", "Epoch 664/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 79.6816 - calc_mre_K: 0.4863 - val_loss: 78.4452 - val_calc_mre_K: 0.4794\n", "Epoch 665/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 79.9532 - calc_mre_K: 0.4880 - val_loss: 92.7975 - val_calc_mre_K: 0.5672\n", "Epoch 666/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.1137 - calc_mre_K: 0.4829 - val_loss: 75.5468 - val_calc_mre_K: 0.4617\n", "Epoch 667/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.5487 - calc_mre_K: 0.4855 - val_loss: 76.1708 - val_calc_mre_K: 0.4656\n", "Epoch 668/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 79.4598 - calc_mre_K: 0.4850 - val_loss: 76.2132 - val_calc_mre_K: 0.4658\n", "Epoch 669/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.3404 - calc_mre_K: 0.4843 - val_loss: 76.2156 - val_calc_mre_K: 0.4658\n", "Epoch 670/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.8430 - calc_mre_K: 0.4812 - val_loss: 90.4779 - val_calc_mre_K: 0.5530\n", "Epoch 671/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.1211 - calc_mre_K: 0.4890 - val_loss: 80.6851 - val_calc_mre_K: 0.4931\n", "Epoch 672/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 78.8271 - calc_mre_K: 0.4811 - val_loss: 77.6186 - val_calc_mre_K: 0.4744\n", "Epoch 673/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 79.1796 - calc_mre_K: 0.4833 - val_loss: 78.7684 - val_calc_mre_K: 0.4814\n", "Epoch 674/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 80.2572 - calc_mre_K: 0.4899 - val_loss: 77.5710 - val_calc_mre_K: 0.4741\n", "Epoch 675/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.8381 - calc_mre_K: 0.4812 - val_loss: 78.9315 - val_calc_mre_K: 0.4824\n", "Epoch 676/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.1674 - calc_mre_K: 0.4832 - val_loss: 84.9555 - val_calc_mre_K: 0.5193\n", "Epoch 677/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 80.1877 - calc_mre_K: 0.4894 - val_loss: 81.7227 - val_calc_mre_K: 0.4995\n", "Epoch 678/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 78.7001 - calc_mre_K: 0.4803 - val_loss: 80.9586 - val_calc_mre_K: 0.4948\n", "Epoch 679/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.6204 - calc_mre_K: 0.4860 - val_loss: 79.4980 - val_calc_mre_K: 0.4859\n", "Epoch 680/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 79.7538 - calc_mre_K: 0.4868 - val_loss: 79.6147 - val_calc_mre_K: 0.4866\n", "Epoch 681/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.1855 - calc_mre_K: 0.4772 - val_loss: 82.6658 - val_calc_mre_K: 0.5052\n", "Epoch 682/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 79.0095 - calc_mre_K: 0.4822 - val_loss: 81.4938 - val_calc_mre_K: 0.4981\n", "Epoch 683/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 79.5130 - calc_mre_K: 0.4853 - val_loss: 78.6661 - val_calc_mre_K: 0.4808\n", "Epoch 684/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 78.6651 - calc_mre_K: 0.4801 - val_loss: 78.2105 - val_calc_mre_K: 0.4780\n", "Epoch 685/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.7097 - calc_mre_K: 0.4804 - val_loss: 78.2422 - val_calc_mre_K: 0.4782\n", "Epoch 686/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 79.4313 - calc_mre_K: 0.4848 - val_loss: 79.1577 - val_calc_mre_K: 0.4838\n", "Epoch 687/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.5887 - calc_mre_K: 0.4858 - val_loss: 74.9157 - val_calc_mre_K: 0.4579\n", "Epoch 688/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.4066 - calc_mre_K: 0.4847 - val_loss: 77.5264 - val_calc_mre_K: 0.4738\n", "Epoch 689/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 79.1942 - calc_mre_K: 0.4834 - val_loss: 79.9662 - val_calc_mre_K: 0.4887\n", "Epoch 690/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.9376 - calc_mre_K: 0.4818 - val_loss: 85.9422 - val_calc_mre_K: 0.5253\n", "Epoch 691/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.7807 - calc_mre_K: 0.4808 - val_loss: 75.6300 - val_calc_mre_K: 0.4622\n", "Epoch 692/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.0589 - calc_mre_K: 0.4764 - val_loss: 85.9689 - val_calc_mre_K: 0.5254\n", "Epoch 693/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.0083 - calc_mre_K: 0.4822 - val_loss: 79.4006 - val_calc_mre_K: 0.4853\n", "Epoch 694/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.7597 - calc_mre_K: 0.4868 - val_loss: 74.9714 - val_calc_mre_K: 0.4582\n", "Epoch 695/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 80.7174 - calc_mre_K: 0.4927 - val_loss: 75.2823 - val_calc_mre_K: 0.4601\n", "Epoch 696/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.8769 - calc_mre_K: 0.4753 - val_loss: 76.9646 - val_calc_mre_K: 0.4704\n", "Epoch 697/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.6615 - calc_mre_K: 0.4740 - val_loss: 84.6240 - val_calc_mre_K: 0.5172\n", "Epoch 698/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.8241 - calc_mre_K: 0.4872 - val_loss: 79.2511 - val_calc_mre_K: 0.4844\n", "Epoch 699/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 79.0223 - calc_mre_K: 0.4823 - val_loss: 76.5035 - val_calc_mre_K: 0.4676\n", "Epoch 700/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.7648 - calc_mre_K: 0.4807 - val_loss: 80.7132 - val_calc_mre_K: 0.4933\n", "Epoch 701/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 78.1678 - calc_mre_K: 0.4771 - val_loss: 76.9234 - val_calc_mre_K: 0.4702\n", "Epoch 702/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.5651 - calc_mre_K: 0.4795 - val_loss: 77.6392 - val_calc_mre_K: 0.4745\n", "Epoch 703/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 78.6606 - calc_mre_K: 0.4801 - val_loss: 76.9037 - val_calc_mre_K: 0.4700\n", "Epoch 704/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.0073 - calc_mre_K: 0.4822 - val_loss: 77.9583 - val_calc_mre_K: 0.4765\n", "Epoch 705/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 78.2324 - calc_mre_K: 0.4775 - val_loss: 81.4888 - val_calc_mre_K: 0.4980\n", "Epoch 706/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.2399 - calc_mre_K: 0.4775 - val_loss: 79.1666 - val_calc_mre_K: 0.4839\n", "Epoch 707/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.7239 - calc_mre_K: 0.4805 - val_loss: 76.0261 - val_calc_mre_K: 0.4647\n", "Epoch 708/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 78.4201 - calc_mre_K: 0.4786 - val_loss: 77.0172 - val_calc_mre_K: 0.4707\n", "Epoch 709/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 78.3979 - calc_mre_K: 0.4785 - val_loss: 81.2832 - val_calc_mre_K: 0.4968\n", "Epoch 710/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.4960 - calc_mre_K: 0.4791 - val_loss: 79.2657 - val_calc_mre_K: 0.4845\n", "Epoch 711/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.6637 - calc_mre_K: 0.4801 - val_loss: 76.1678 - val_calc_mre_K: 0.4655\n", "Epoch 712/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 78.1736 - calc_mre_K: 0.4771 - val_loss: 77.9545 - val_calc_mre_K: 0.4764\n", "Epoch 713/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.2781 - calc_mre_K: 0.4778 - val_loss: 79.2701 - val_calc_mre_K: 0.4845\n", "Epoch 714/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 78.0192 - calc_mre_K: 0.4762 - val_loss: 80.8515 - val_calc_mre_K: 0.4941\n", "Epoch 715/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.6416 - calc_mre_K: 0.4800 - val_loss: 74.2747 - val_calc_mre_K: 0.4539\n", "Epoch 716/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.6817 - calc_mre_K: 0.4741 - val_loss: 80.3660 - val_calc_mre_K: 0.4912\n", "Epoch 717/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.9744 - calc_mre_K: 0.4820 - val_loss: 74.8350 - val_calc_mre_K: 0.4574\n", "Epoch 718/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.5253 - calc_mre_K: 0.4793 - val_loss: 77.8145 - val_calc_mre_K: 0.4756\n", "Epoch 719/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.3642 - calc_mre_K: 0.4783 - val_loss: 75.3179 - val_calc_mre_K: 0.4603\n", "Epoch 720/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.8681 - calc_mre_K: 0.4814 - val_loss: 85.2466 - val_calc_mre_K: 0.5210\n", "Epoch 721/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.1541 - calc_mre_K: 0.4770 - val_loss: 83.3928 - val_calc_mre_K: 0.5097\n", "Epoch 722/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 77.4030 - calc_mre_K: 0.4724 - val_loss: 76.6117 - val_calc_mre_K: 0.4682\n", "Epoch 723/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.1465 - calc_mre_K: 0.4770 - val_loss: 80.4226 - val_calc_mre_K: 0.4915\n", "Epoch 724/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.3843 - calc_mre_K: 0.4784 - val_loss: 80.8648 - val_calc_mre_K: 0.4942\n", "Epoch 725/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 78.3181 - calc_mre_K: 0.4780 - val_loss: 73.9927 - val_calc_mre_K: 0.4522\n", "Epoch 726/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.0761 - calc_mre_K: 0.4765 - val_loss: 80.8112 - val_calc_mre_K: 0.4939\n", "Epoch 727/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 58us/step - loss: 78.6627 - calc_mre_K: 0.4801 - val_loss: 84.8342 - val_calc_mre_K: 0.5185\n", "Epoch 728/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 77.9144 - calc_mre_K: 0.4756 - val_loss: 80.8966 - val_calc_mre_K: 0.4944\n", "Epoch 729/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.9505 - calc_mre_K: 0.4758 - val_loss: 73.6831 - val_calc_mre_K: 0.4503\n", "Epoch 730/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 78.3226 - calc_mre_K: 0.4780 - val_loss: 75.4109 - val_calc_mre_K: 0.4609\n", "Epoch 731/2000\n", "48000/48000 [==============================] - 3s 64us/step - loss: 78.4290 - calc_mre_K: 0.4787 - val_loss: 75.1637 - val_calc_mre_K: 0.4594\n", "Epoch 732/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.0997 - calc_mre_K: 0.4767 - val_loss: 77.6759 - val_calc_mre_K: 0.4747\n", "Epoch 733/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 78.4055 - calc_mre_K: 0.4785 - val_loss: 86.0408 - val_calc_mre_K: 0.5259\n", "Epoch 734/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 78.8260 - calc_mre_K: 0.4811 - val_loss: 83.2825 - val_calc_mre_K: 0.5090\n", "Epoch 735/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.9157 - calc_mre_K: 0.4756 - val_loss: 73.8278 - val_calc_mre_K: 0.4512\n", "Epoch 736/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 78.5713 - calc_mre_K: 0.4796 - val_loss: 80.0016 - val_calc_mre_K: 0.4890\n", "Epoch 737/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 77.8274 - calc_mre_K: 0.4750 - val_loss: 78.0546 - val_calc_mre_K: 0.4771\n", "Epoch 738/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 79.0235 - calc_mre_K: 0.4823 - val_loss: 77.8336 - val_calc_mre_K: 0.4757\n", "Epoch 739/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.1610 - calc_mre_K: 0.4771 - val_loss: 82.9839 - val_calc_mre_K: 0.5072\n", "Epoch 740/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 78.5859 - calc_mre_K: 0.4797 - val_loss: 75.9789 - val_calc_mre_K: 0.4644\n", "Epoch 741/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.7600 - calc_mre_K: 0.4746 - val_loss: 80.8072 - val_calc_mre_K: 0.4939\n", "Epoch 742/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.5797 - calc_mre_K: 0.4735 - val_loss: 90.8631 - val_calc_mre_K: 0.5553\n", "Epoch 743/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 78.1394 - calc_mre_K: 0.4769 - val_loss: 75.6425 - val_calc_mre_K: 0.4623\n", "Epoch 744/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.5263 - calc_mre_K: 0.4793 - val_loss: 79.4230 - val_calc_mre_K: 0.4854\n", "Epoch 745/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.7750 - calc_mre_K: 0.4747 - val_loss: 76.7254 - val_calc_mre_K: 0.4689\n", "Epoch 746/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.6887 - calc_mre_K: 0.4803 - val_loss: 86.0469 - val_calc_mre_K: 0.5259\n", "Epoch 747/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.9393 - calc_mre_K: 0.4696 - val_loss: 74.3626 - val_calc_mre_K: 0.4545\n", "Epoch 748/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 79.6302 - calc_mre_K: 0.4860 - val_loss: 74.8924 - val_calc_mre_K: 0.4577\n", "Epoch 749/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.6170 - calc_mre_K: 0.4737 - val_loss: 73.3312 - val_calc_mre_K: 0.4482\n", "Epoch 750/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.4375 - calc_mre_K: 0.4726 - val_loss: 76.2646 - val_calc_mre_K: 0.4661\n", "Epoch 751/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 77.7256 - calc_mre_K: 0.4744 - val_loss: 77.3512 - val_calc_mre_K: 0.4728\n", "Epoch 752/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 77.6209 - calc_mre_K: 0.4738 - val_loss: 75.4514 - val_calc_mre_K: 0.4612\n", "Epoch 753/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 77.3630 - calc_mre_K: 0.4722 - val_loss: 78.2941 - val_calc_mre_K: 0.4785\n", "Epoch 754/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.2864 - calc_mre_K: 0.4778 - val_loss: 77.1507 - val_calc_mre_K: 0.4715\n", "Epoch 755/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.4076 - calc_mre_K: 0.4786 - val_loss: 82.4055 - val_calc_mre_K: 0.5036\n", "Epoch 756/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 77.5873 - calc_mre_K: 0.4736 - val_loss: 76.0212 - val_calc_mre_K: 0.4646\n", "Epoch 757/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.8938 - calc_mre_K: 0.4754 - val_loss: 74.4580 - val_calc_mre_K: 0.4551\n", "Epoch 758/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 77.5975 - calc_mre_K: 0.4736 - val_loss: 76.0035 - val_calc_mre_K: 0.4645\n", "Epoch 759/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.6100 - calc_mre_K: 0.4737 - val_loss: 80.2037 - val_calc_mre_K: 0.4902\n", "Epoch 760/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 77.8414 - calc_mre_K: 0.4751 - val_loss: 74.2484 - val_calc_mre_K: 0.4538\n", "Epoch 761/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 78.3436 - calc_mre_K: 0.4782 - val_loss: 74.0830 - val_calc_mre_K: 0.4528\n", "Epoch 762/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 77.8018 - calc_mre_K: 0.4749 - val_loss: 79.8509 - val_calc_mre_K: 0.4880\n", "Epoch 763/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.2587 - calc_mre_K: 0.4715 - val_loss: 75.9327 - val_calc_mre_K: 0.4641\n", "Epoch 764/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.8985 - calc_mre_K: 0.4755 - val_loss: 77.5176 - val_calc_mre_K: 0.4738\n", "Epoch 765/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 77.6902 - calc_mre_K: 0.4742 - val_loss: 82.9288 - val_calc_mre_K: 0.5068\n", "Epoch 766/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 77.6309 - calc_mre_K: 0.4738 - val_loss: 77.2135 - val_calc_mre_K: 0.4719\n", "Epoch 767/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 78.1026 - calc_mre_K: 0.4767 - val_loss: 75.0419 - val_calc_mre_K: 0.4586\n", "Epoch 768/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.3088 - calc_mre_K: 0.4719 - val_loss: 77.1916 - val_calc_mre_K: 0.4718\n", "Epoch 769/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.1160 - calc_mre_K: 0.4707 - val_loss: 75.2903 - val_calc_mre_K: 0.4602\n", "Epoch 770/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.5115 - calc_mre_K: 0.4731 - val_loss: 80.7306 - val_calc_mre_K: 0.4934\n", "Epoch 771/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 78.0542 - calc_mre_K: 0.4764 - val_loss: 75.9133 - val_calc_mre_K: 0.4640\n", "Epoch 772/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.5382 - calc_mre_K: 0.4733 - val_loss: 77.8056 - val_calc_mre_K: 0.4755\n", "Epoch 773/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.0630 - calc_mre_K: 0.4704 - val_loss: 77.9714 - val_calc_mre_K: 0.4766\n", "Epoch 774/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.2025 - calc_mre_K: 0.4712 - val_loss: 75.2449 - val_calc_mre_K: 0.4599\n", "Epoch 775/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 78.5270 - calc_mre_K: 0.4793 - val_loss: 79.6337 - val_calc_mre_K: 0.4867\n", "Epoch 776/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.0441 - calc_mre_K: 0.4702 - val_loss: 76.9449 - val_calc_mre_K: 0.4703\n", "Epoch 777/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.4204 - calc_mre_K: 0.4664 - val_loss: 76.1187 - val_calc_mre_K: 0.4652\n", "Epoch 778/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.7771 - calc_mre_K: 0.4747 - val_loss: 78.8999 - val_calc_mre_K: 0.4822\n", "Epoch 779/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.4558 - calc_mre_K: 0.4728 - val_loss: 77.8423 - val_calc_mre_K: 0.4758\n", "Epoch 780/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 77.6450 - calc_mre_K: 0.4739 - val_loss: 79.5316 - val_calc_mre_K: 0.4861\n", "Epoch 781/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 78.0095 - calc_mre_K: 0.4761 - val_loss: 76.7875 - val_calc_mre_K: 0.4693\n", "Epoch 782/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.2581 - calc_mre_K: 0.4715 - val_loss: 73.9897 - val_calc_mre_K: 0.4522\n", "Epoch 783/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 78.4435 - calc_mre_K: 0.4788 - val_loss: 78.0770 - val_calc_mre_K: 0.4772\n", "Epoch 784/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.4669 - calc_mre_K: 0.4667 - val_loss: 84.1983 - val_calc_mre_K: 0.5146\n", "Epoch 785/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 78.2151 - calc_mre_K: 0.4774 - val_loss: 76.7049 - val_calc_mre_K: 0.4688\n", "Epoch 786/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 76.7210 - calc_mre_K: 0.4683 - val_loss: 76.2578 - val_calc_mre_K: 0.4661\n", "Epoch 787/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.1319 - calc_mre_K: 0.4708 - val_loss: 83.8492 - val_calc_mre_K: 0.5125\n", "Epoch 788/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.2061 - calc_mre_K: 0.4712 - val_loss: 76.6518 - val_calc_mre_K: 0.4685\n", "Epoch 789/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.2650 - calc_mre_K: 0.4716 - val_loss: 75.6030 - val_calc_mre_K: 0.4621\n", "Epoch 790/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 77.0629 - calc_mre_K: 0.4704 - val_loss: 76.0557 - val_calc_mre_K: 0.4648\n", "Epoch 791/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.8600 - calc_mre_K: 0.4752 - val_loss: 80.9513 - val_calc_mre_K: 0.4947\n", "Epoch 792/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.7062 - calc_mre_K: 0.4743 - val_loss: 76.3339 - val_calc_mre_K: 0.4665\n", "Epoch 793/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.3388 - calc_mre_K: 0.4659 - val_loss: 82.1076 - val_calc_mre_K: 0.5018\n", "Epoch 794/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.4720 - calc_mre_K: 0.4729 - val_loss: 73.0964 - val_calc_mre_K: 0.4467\n", "Epoch 795/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 77.2243 - calc_mre_K: 0.4713 - val_loss: 75.7339 - val_calc_mre_K: 0.4629\n", "Epoch 796/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.0279 - calc_mre_K: 0.4701 - val_loss: 74.9003 - val_calc_mre_K: 0.4578\n", "Epoch 797/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 77.0034 - calc_mre_K: 0.4700 - val_loss: 77.5636 - val_calc_mre_K: 0.4741\n", "Epoch 798/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.9633 - calc_mre_K: 0.4697 - val_loss: 73.0600 - val_calc_mre_K: 0.4465\n", "Epoch 799/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.6251 - calc_mre_K: 0.4677 - val_loss: 74.9767 - val_calc_mre_K: 0.4582\n", "Epoch 800/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 77.6696 - calc_mre_K: 0.4741 - val_loss: 79.5865 - val_calc_mre_K: 0.4864\n", "Epoch 801/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 76.8021 - calc_mre_K: 0.4688 - val_loss: 75.8691 - val_calc_mre_K: 0.4637\n", "Epoch 802/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 77.2434 - calc_mre_K: 0.4715 - val_loss: 72.6447 - val_calc_mre_K: 0.4440\n", "Epoch 803/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.9150 - calc_mre_K: 0.4695 - val_loss: 76.8125 - val_calc_mre_K: 0.4695\n", "Epoch 804/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.0433 - calc_mre_K: 0.4702 - val_loss: 73.8261 - val_calc_mre_K: 0.4512\n", "Epoch 805/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.2372 - calc_mre_K: 0.4714 - val_loss: 74.7099 - val_calc_mre_K: 0.4566\n", "Epoch 806/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.5765 - calc_mre_K: 0.4674 - val_loss: 77.8128 - val_calc_mre_K: 0.4756\n", "Epoch 807/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 77.7196 - calc_mre_K: 0.4744 - val_loss: 81.8107 - val_calc_mre_K: 0.5000\n", "Epoch 808/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.0043 - calc_mre_K: 0.4700 - val_loss: 79.1323 - val_calc_mre_K: 0.4836\n", "Epoch 809/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 76.9548 - calc_mre_K: 0.4697 - val_loss: 72.7392 - val_calc_mre_K: 0.4446\n", "Epoch 810/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.3209 - calc_mre_K: 0.4719 - val_loss: 80.3253 - val_calc_mre_K: 0.4909\n", "Epoch 811/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.8674 - calc_mre_K: 0.4692 - val_loss: 73.4790 - val_calc_mre_K: 0.4491\n", "Epoch 812/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.5628 - calc_mre_K: 0.4673 - val_loss: 73.1358 - val_calc_mre_K: 0.4470\n", "Epoch 813/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 76.8508 - calc_mre_K: 0.4691 - val_loss: 77.6272 - val_calc_mre_K: 0.4744\n", "Epoch 814/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.3719 - calc_mre_K: 0.4661 - val_loss: 79.3792 - val_calc_mre_K: 0.4852\n", "Epoch 815/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.3969 - calc_mre_K: 0.4724 - val_loss: 77.3930 - val_calc_mre_K: 0.4730\n", "Epoch 816/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.7089 - calc_mre_K: 0.4682 - val_loss: 74.6599 - val_calc_mre_K: 0.4563\n", "Epoch 817/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 76.6768 - calc_mre_K: 0.4680 - val_loss: 75.6664 - val_calc_mre_K: 0.4625\n", "Epoch 818/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.8363 - calc_mre_K: 0.4690 - val_loss: 73.7362 - val_calc_mre_K: 0.4507\n", "Epoch 819/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 76.9292 - calc_mre_K: 0.4695 - val_loss: 76.5610 - val_calc_mre_K: 0.4679\n", "Epoch 820/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 76.6734 - calc_mre_K: 0.4680 - val_loss: 76.1383 - val_calc_mre_K: 0.4653\n", "Epoch 821/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.8588 - calc_mre_K: 0.4691 - val_loss: 75.1310 - val_calc_mre_K: 0.4592\n", "Epoch 822/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 76.7074 - calc_mre_K: 0.4682 - val_loss: 77.7053 - val_calc_mre_K: 0.4749\n", "Epoch 823/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 77.2159 - calc_mre_K: 0.4713 - val_loss: 74.0998 - val_calc_mre_K: 0.4529\n", "Epoch 824/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.7894 - calc_mre_K: 0.4687 - val_loss: 74.5837 - val_calc_mre_K: 0.4558\n", "Epoch 825/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.5918 - calc_mre_K: 0.4675 - val_loss: 75.8626 - val_calc_mre_K: 0.4637\n", "Epoch 826/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.6919 - calc_mre_K: 0.4681 - val_loss: 76.1804 - val_calc_mre_K: 0.4656\n", "Epoch 827/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.6308 - calc_mre_K: 0.4677 - val_loss: 73.3450 - val_calc_mre_K: 0.4483\n", "Epoch 828/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.7958 - calc_mre_K: 0.4687 - val_loss: 74.9832 - val_calc_mre_K: 0.4583\n", "Epoch 829/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.8817 - calc_mre_K: 0.4692 - val_loss: 80.0781 - val_calc_mre_K: 0.4894\n", "Epoch 830/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.7558 - calc_mre_K: 0.4685 - val_loss: 73.7853 - val_calc_mre_K: 0.4510\n", "Epoch 831/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 55us/step - loss: 76.7612 - calc_mre_K: 0.4685 - val_loss: 74.3267 - val_calc_mre_K: 0.4543\n", "Epoch 832/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.7138 - calc_mre_K: 0.4682 - val_loss: 77.8206 - val_calc_mre_K: 0.4756\n", "Epoch 833/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.5893 - calc_mre_K: 0.4675 - val_loss: 79.3091 - val_calc_mre_K: 0.4847\n", "Epoch 834/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 77.0921 - calc_mre_K: 0.4705 - val_loss: 83.1803 - val_calc_mre_K: 0.5084\n", "Epoch 835/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 76.5175 - calc_mre_K: 0.4670 - val_loss: 82.1222 - val_calc_mre_K: 0.5019\n", "Epoch 836/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.1910 - calc_mre_K: 0.4711 - val_loss: 76.9383 - val_calc_mre_K: 0.4702\n", "Epoch 837/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 77.2267 - calc_mre_K: 0.4714 - val_loss: 76.9687 - val_calc_mre_K: 0.4704\n", "Epoch 838/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.9634 - calc_mre_K: 0.4636 - val_loss: 82.8781 - val_calc_mre_K: 0.5065\n", "Epoch 839/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.9295 - calc_mre_K: 0.4695 - val_loss: 73.4074 - val_calc_mre_K: 0.4487\n", "Epoch 840/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.6701 - calc_mre_K: 0.4680 - val_loss: 79.8560 - val_calc_mre_K: 0.4881\n", "Epoch 841/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 77.6095 - calc_mre_K: 0.4737 - val_loss: 78.2259 - val_calc_mre_K: 0.4781\n", "Epoch 842/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.7363 - calc_mre_K: 0.4684 - val_loss: 73.9550 - val_calc_mre_K: 0.4520\n", "Epoch 843/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.9661 - calc_mre_K: 0.4637 - val_loss: 76.0434 - val_calc_mre_K: 0.4648\n", "Epoch 844/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 76.6781 - calc_mre_K: 0.4680 - val_loss: 74.5330 - val_calc_mre_K: 0.4555\n", "Epoch 845/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 77.0743 - calc_mre_K: 0.4704 - val_loss: 77.7429 - val_calc_mre_K: 0.4751\n", "Epoch 846/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 76.2876 - calc_mre_K: 0.4656 - val_loss: 78.4469 - val_calc_mre_K: 0.4795\n", "Epoch 847/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.9068 - calc_mre_K: 0.4633 - val_loss: 75.1117 - val_calc_mre_K: 0.4591\n", "Epoch 848/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 76.2157 - calc_mre_K: 0.4652 - val_loss: 74.5479 - val_calc_mre_K: 0.4556\n", "Epoch 849/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 76.4971 - calc_mre_K: 0.4669 - val_loss: 75.6664 - val_calc_mre_K: 0.4625\n", "Epoch 850/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.9865 - calc_mre_K: 0.4638 - val_loss: 77.0451 - val_calc_mre_K: 0.4709\n", "Epoch 851/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 76.2991 - calc_mre_K: 0.4657 - val_loss: 73.8944 - val_calc_mre_K: 0.4516\n", "Epoch 852/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 76.9880 - calc_mre_K: 0.4699 - val_loss: 73.5337 - val_calc_mre_K: 0.4494\n", "Epoch 853/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 76.9204 - calc_mre_K: 0.4695 - val_loss: 75.5672 - val_calc_mre_K: 0.4618\n", "Epoch 854/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.1211 - calc_mre_K: 0.4646 - val_loss: 74.2494 - val_calc_mre_K: 0.4538\n", "Epoch 855/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.6719 - calc_mre_K: 0.4680 - val_loss: 73.8108 - val_calc_mre_K: 0.4511\n", "Epoch 856/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 76.1992 - calc_mre_K: 0.4651 - val_loss: 73.3622 - val_calc_mre_K: 0.4484\n", "Epoch 857/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.9962 - calc_mre_K: 0.4638 - val_loss: 85.2367 - val_calc_mre_K: 0.5209\n", "Epoch 858/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 76.1798 - calc_mre_K: 0.4650 - val_loss: 79.5255 - val_calc_mre_K: 0.4860\n", "Epoch 859/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.4925 - calc_mre_K: 0.4669 - val_loss: 75.3295 - val_calc_mre_K: 0.4604\n", "Epoch 860/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.2845 - calc_mre_K: 0.4656 - val_loss: 74.9648 - val_calc_mre_K: 0.4582\n", "Epoch 861/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.1168 - calc_mre_K: 0.4646 - val_loss: 73.4049 - val_calc_mre_K: 0.4486\n", "Epoch 862/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.4714 - calc_mre_K: 0.4667 - val_loss: 73.6728 - val_calc_mre_K: 0.4503\n", "Epoch 863/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.8534 - calc_mre_K: 0.4630 - val_loss: 73.5419 - val_calc_mre_K: 0.4495\n", "Epoch 864/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.4094 - calc_mre_K: 0.4664 - val_loss: 80.0106 - val_calc_mre_K: 0.4890\n", "Epoch 865/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.7468 - calc_mre_K: 0.4684 - val_loss: 73.2823 - val_calc_mre_K: 0.4479\n", "Epoch 866/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.7539 - calc_mre_K: 0.4624 - val_loss: 78.4554 - val_calc_mre_K: 0.4795\n", "Epoch 867/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 76.3511 - calc_mre_K: 0.4660 - val_loss: 73.7789 - val_calc_mre_K: 0.4509\n", "Epoch 868/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.8690 - calc_mre_K: 0.4631 - val_loss: 73.1468 - val_calc_mre_K: 0.4471\n", "Epoch 869/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.7140 - calc_mre_K: 0.4682 - val_loss: 73.0188 - val_calc_mre_K: 0.4463\n", "Epoch 870/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.4086 - calc_mre_K: 0.4603 - val_loss: 76.3666 - val_calc_mre_K: 0.4667\n", "Epoch 871/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.8992 - calc_mre_K: 0.4633 - val_loss: 74.8066 - val_calc_mre_K: 0.4572\n", "Epoch 872/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.1653 - calc_mre_K: 0.4649 - val_loss: 75.0184 - val_calc_mre_K: 0.4585\n", "Epoch 873/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.0547 - calc_mre_K: 0.4642 - val_loss: 73.0752 - val_calc_mre_K: 0.4466\n", "Epoch 874/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.3755 - calc_mre_K: 0.4662 - val_loss: 79.8924 - val_calc_mre_K: 0.4883\n", "Epoch 875/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.3559 - calc_mre_K: 0.4660 - val_loss: 74.6050 - val_calc_mre_K: 0.4560\n", "Epoch 876/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.1912 - calc_mre_K: 0.4650 - val_loss: 75.3396 - val_calc_mre_K: 0.4605\n", "Epoch 877/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.6913 - calc_mre_K: 0.4620 - val_loss: 73.4236 - val_calc_mre_K: 0.4487\n", "Epoch 878/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.6181 - calc_mre_K: 0.4676 - val_loss: 74.5920 - val_calc_mre_K: 0.4559\n", "Epoch 879/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.2014 - calc_mre_K: 0.4651 - val_loss: 78.8734 - val_calc_mre_K: 0.4821\n", "Epoch 880/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.0899 - calc_mre_K: 0.4644 - val_loss: 72.3771 - val_calc_mre_K: 0.4423\n", "Epoch 881/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.8085 - calc_mre_K: 0.4627 - val_loss: 74.0301 - val_calc_mre_K: 0.4525\n", "Epoch 882/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 77.1431 - calc_mre_K: 0.4708 - val_loss: 83.2413 - val_calc_mre_K: 0.5087\n", "Epoch 883/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.6575 - calc_mre_K: 0.4618 - val_loss: 73.1610 - val_calc_mre_K: 0.4472\n", "Epoch 884/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.7800 - calc_mre_K: 0.4625 - val_loss: 74.9881 - val_calc_mre_K: 0.4583\n", "Epoch 885/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.5018 - calc_mre_K: 0.4608 - val_loss: 72.6429 - val_calc_mre_K: 0.4440\n", "Epoch 886/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.0838 - calc_mre_K: 0.4644 - val_loss: 77.5472 - val_calc_mre_K: 0.4740\n", "Epoch 887/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 75.6518 - calc_mre_K: 0.4617 - val_loss: 75.7267 - val_calc_mre_K: 0.4628\n", "Epoch 888/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.7121 - calc_mre_K: 0.4621 - val_loss: 77.2829 - val_calc_mre_K: 0.4723\n", "Epoch 889/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.8115 - calc_mre_K: 0.4627 - val_loss: 78.0226 - val_calc_mre_K: 0.4768\n", "Epoch 890/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.0737 - calc_mre_K: 0.4582 - val_loss: 72.7438 - val_calc_mre_K: 0.4446\n", "Epoch 891/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.6920 - calc_mre_K: 0.4681 - val_loss: 78.1002 - val_calc_mre_K: 0.4773\n", "Epoch 892/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.2501 - calc_mre_K: 0.4593 - val_loss: 81.9437 - val_calc_mre_K: 0.5008\n", "Epoch 893/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 76.0491 - calc_mre_K: 0.4642 - val_loss: 73.4678 - val_calc_mre_K: 0.4490\n", "Epoch 894/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.8122 - calc_mre_K: 0.4627 - val_loss: 79.6387 - val_calc_mre_K: 0.4867\n", "Epoch 895/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.4257 - calc_mre_K: 0.4604 - val_loss: 81.5580 - val_calc_mre_K: 0.4984\n", "Epoch 896/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.8920 - calc_mre_K: 0.4632 - val_loss: 75.5214 - val_calc_mre_K: 0.4616\n", "Epoch 897/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.5323 - calc_mre_K: 0.4610 - val_loss: 76.2468 - val_calc_mre_K: 0.4660\n", "Epoch 898/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.6858 - calc_mre_K: 0.4619 - val_loss: 73.5163 - val_calc_mre_K: 0.4493\n", "Epoch 899/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.9528 - calc_mre_K: 0.4636 - val_loss: 75.6460 - val_calc_mre_K: 0.4623\n", "Epoch 900/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.2655 - calc_mre_K: 0.4594 - val_loss: 74.8842 - val_calc_mre_K: 0.4577\n", "Epoch 901/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.7907 - calc_mre_K: 0.4626 - val_loss: 75.2264 - val_calc_mre_K: 0.4598\n", "Epoch 902/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 75.6279 - calc_mre_K: 0.4616 - val_loss: 79.0788 - val_calc_mre_K: 0.4833\n", "Epoch 903/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.8417 - calc_mre_K: 0.4629 - val_loss: 76.2286 - val_calc_mre_K: 0.4659\n", "Epoch 904/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.1667 - calc_mre_K: 0.4588 - val_loss: 72.4379 - val_calc_mre_K: 0.4427\n", "Epoch 905/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 76.6725 - calc_mre_K: 0.4680 - val_loss: 77.6656 - val_calc_mre_K: 0.4747\n", "Epoch 906/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.5542 - calc_mre_K: 0.4611 - val_loss: 72.7256 - val_calc_mre_K: 0.4445\n", "Epoch 907/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.4963 - calc_mre_K: 0.4608 - val_loss: 76.4816 - val_calc_mre_K: 0.4674\n", "Epoch 908/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.0045 - calc_mre_K: 0.4639 - val_loss: 74.4465 - val_calc_mre_K: 0.4550\n", "Epoch 909/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.7153 - calc_mre_K: 0.4621 - val_loss: 75.8705 - val_calc_mre_K: 0.4637\n", "Epoch 910/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.2789 - calc_mre_K: 0.4595 - val_loss: 74.2200 - val_calc_mre_K: 0.4536\n", "Epoch 911/2000\n", "48000/48000 [==============================] - 3s 63us/step - loss: 75.2773 - calc_mre_K: 0.4595 - val_loss: 74.0391 - val_calc_mre_K: 0.4525\n", "Epoch 912/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.7080 - calc_mre_K: 0.4621 - val_loss: 78.2615 - val_calc_mre_K: 0.4783\n", "Epoch 913/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.7241 - calc_mre_K: 0.4622 - val_loss: 73.6820 - val_calc_mre_K: 0.4503\n", "Epoch 914/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.5670 - calc_mre_K: 0.4612 - val_loss: 74.6677 - val_calc_mre_K: 0.4563\n", "Epoch 915/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 76.2355 - calc_mre_K: 0.4653 - val_loss: 82.1596 - val_calc_mre_K: 0.5021\n", "Epoch 916/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 76.0292 - calc_mre_K: 0.4640 - val_loss: 83.1513 - val_calc_mre_K: 0.5082\n", "Epoch 917/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.1021 - calc_mre_K: 0.4584 - val_loss: 73.6640 - val_calc_mre_K: 0.4502\n", "Epoch 918/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.1358 - calc_mre_K: 0.4647 - val_loss: 79.3723 - val_calc_mre_K: 0.4851\n", "Epoch 919/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.1657 - calc_mre_K: 0.4588 - val_loss: 78.3152 - val_calc_mre_K: 0.4786\n", "Epoch 920/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.9835 - calc_mre_K: 0.4638 - val_loss: 76.1738 - val_calc_mre_K: 0.4655\n", "Epoch 921/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 75.2897 - calc_mre_K: 0.4595 - val_loss: 74.5351 - val_calc_mre_K: 0.4555\n", "Epoch 922/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.0215 - calc_mre_K: 0.4579 - val_loss: 76.1965 - val_calc_mre_K: 0.4657\n", "Epoch 923/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 76.2249 - calc_mre_K: 0.4652 - val_loss: 100.2480 - val_calc_mre_K: 0.6127\n", "Epoch 924/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.9188 - calc_mre_K: 0.4634 - val_loss: 73.5489 - val_calc_mre_K: 0.4495\n", "Epoch 925/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 74.8621 - calc_mre_K: 0.4569 - val_loss: 75.9934 - val_calc_mre_K: 0.4644\n", "Epoch 926/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.6708 - calc_mre_K: 0.4680 - val_loss: 77.5002 - val_calc_mre_K: 0.4737\n", "Epoch 927/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.2764 - calc_mre_K: 0.4595 - val_loss: 77.1200 - val_calc_mre_K: 0.4713\n", "Epoch 928/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 76.0915 - calc_mre_K: 0.4644 - val_loss: 73.0056 - val_calc_mre_K: 0.4462\n", "Epoch 929/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 74.8964 - calc_mre_K: 0.4571 - val_loss: 77.4249 - val_calc_mre_K: 0.4732\n", "Epoch 930/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.8503 - calc_mre_K: 0.4630 - val_loss: 75.6184 - val_calc_mre_K: 0.4621\n", "Epoch 931/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.4830 - calc_mre_K: 0.4546 - val_loss: 74.0860 - val_calc_mre_K: 0.4528\n", "Epoch 932/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.4200 - calc_mre_K: 0.4603 - val_loss: 76.8052 - val_calc_mre_K: 0.4694\n", "Epoch 933/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.8584 - calc_mre_K: 0.4569 - val_loss: 74.1815 - val_calc_mre_K: 0.4534\n", "Epoch 934/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.6551 - calc_mre_K: 0.4618 - val_loss: 72.2374 - val_calc_mre_K: 0.4415\n", "Epoch 935/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 52us/step - loss: 75.3706 - calc_mre_K: 0.4600 - val_loss: 74.5149 - val_calc_mre_K: 0.4554\n", "Epoch 936/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.2933 - calc_mre_K: 0.4596 - val_loss: 76.3564 - val_calc_mre_K: 0.4667\n", "Epoch 937/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.4620 - calc_mre_K: 0.4606 - val_loss: 78.1654 - val_calc_mre_K: 0.4777\n", "Epoch 938/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.5151 - calc_mre_K: 0.4609 - val_loss: 76.8468 - val_calc_mre_K: 0.4697\n", "Epoch 939/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.3785 - calc_mre_K: 0.4601 - val_loss: 73.0442 - val_calc_mre_K: 0.4464\n", "Epoch 940/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.3814 - calc_mre_K: 0.4601 - val_loss: 74.2858 - val_calc_mre_K: 0.4540\n", "Epoch 941/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.5647 - calc_mre_K: 0.4612 - val_loss: 75.7866 - val_calc_mre_K: 0.4632\n", "Epoch 942/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.9374 - calc_mre_K: 0.4574 - val_loss: 80.0542 - val_calc_mre_K: 0.4893\n", "Epoch 943/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 75.6481 - calc_mre_K: 0.4617 - val_loss: 72.0878 - val_calc_mre_K: 0.4406\n", "Epoch 944/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 74.7897 - calc_mre_K: 0.4565 - val_loss: 75.9984 - val_calc_mre_K: 0.4645\n", "Epoch 945/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 74.8958 - calc_mre_K: 0.4571 - val_loss: 82.9698 - val_calc_mre_K: 0.5071\n", "Epoch 946/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.4556 - calc_mre_K: 0.4605 - val_loss: 72.3099 - val_calc_mre_K: 0.4419\n", "Epoch 947/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.0603 - calc_mre_K: 0.4581 - val_loss: 79.0405 - val_calc_mre_K: 0.4831\n", "Epoch 948/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.7118 - calc_mre_K: 0.4621 - val_loss: 74.7035 - val_calc_mre_K: 0.4566\n", "Epoch 949/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 74.5820 - calc_mre_K: 0.4552 - val_loss: 76.6418 - val_calc_mre_K: 0.4684\n", "Epoch 950/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.2263 - calc_mre_K: 0.4591 - val_loss: 73.8007 - val_calc_mre_K: 0.4510\n", "Epoch 951/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.4148 - calc_mre_K: 0.4603 - val_loss: 72.5722 - val_calc_mre_K: 0.4435\n", "Epoch 952/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.3442 - calc_mre_K: 0.4538 - val_loss: 77.1052 - val_calc_mre_K: 0.4712\n", "Epoch 953/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 74.7309 - calc_mre_K: 0.4561 - val_loss: 72.5974 - val_calc_mre_K: 0.4437\n", "Epoch 954/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.0423 - calc_mre_K: 0.4580 - val_loss: 73.6938 - val_calc_mre_K: 0.4504\n", "Epoch 955/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 75.2143 - calc_mre_K: 0.4591 - val_loss: 75.3578 - val_calc_mre_K: 0.4606\n", "Epoch 956/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.1252 - calc_mre_K: 0.4585 - val_loss: 73.9302 - val_calc_mre_K: 0.4518\n", "Epoch 957/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.9015 - calc_mre_K: 0.4572 - val_loss: 81.6304 - val_calc_mre_K: 0.4989\n", "Epoch 958/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.5271 - calc_mre_K: 0.4610 - val_loss: 78.6070 - val_calc_mre_K: 0.4804\n", "Epoch 959/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.6689 - calc_mre_K: 0.4557 - val_loss: 73.5755 - val_calc_mre_K: 0.4497\n", "Epoch 960/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 75.2634 - calc_mre_K: 0.4594 - val_loss: 72.9663 - val_calc_mre_K: 0.4459\n", "Epoch 961/2000\n", "48000/48000 [==============================] - 2s 49us/step - loss: 74.4889 - calc_mre_K: 0.4546 - val_loss: 73.9589 - val_calc_mre_K: 0.4520\n", "Epoch 962/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 75.3128 - calc_mre_K: 0.4597 - val_loss: 73.6949 - val_calc_mre_K: 0.4504\n", "Epoch 963/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 74.6917 - calc_mre_K: 0.4559 - val_loss: 77.9704 - val_calc_mre_K: 0.4765\n", "Epoch 964/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.7379 - calc_mre_K: 0.4562 - val_loss: 71.3339 - val_calc_mre_K: 0.4360\n", "Epoch 965/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.2681 - calc_mre_K: 0.4533 - val_loss: 75.0494 - val_calc_mre_K: 0.4587\n", "Epoch 966/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 75.2719 - calc_mre_K: 0.4594 - val_loss: 72.6327 - val_calc_mre_K: 0.4439\n", "Epoch 967/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.8855 - calc_mre_K: 0.4571 - val_loss: 85.1207 - val_calc_mre_K: 0.5202\n", "Epoch 968/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 75.1010 - calc_mre_K: 0.4584 - val_loss: 74.2757 - val_calc_mre_K: 0.4539\n", "Epoch 969/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.7992 - calc_mre_K: 0.4565 - val_loss: 73.5668 - val_calc_mre_K: 0.4496\n", "Epoch 970/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 75.0155 - calc_mre_K: 0.4579 - val_loss: 73.2361 - val_calc_mre_K: 0.4476\n", "Epoch 971/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.8175 - calc_mre_K: 0.4566 - val_loss: 72.8771 - val_calc_mre_K: 0.4454\n", "Epoch 972/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.4274 - calc_mre_K: 0.4543 - val_loss: 76.1021 - val_calc_mre_K: 0.4651\n", "Epoch 973/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.9257 - calc_mre_K: 0.4573 - val_loss: 79.4508 - val_calc_mre_K: 0.4856\n", "Epoch 974/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.3821 - calc_mre_K: 0.4601 - val_loss: 74.8533 - val_calc_mre_K: 0.4575\n", "Epoch 975/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.9723 - calc_mre_K: 0.4576 - val_loss: 73.7816 - val_calc_mre_K: 0.4509\n", "Epoch 976/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 73.9744 - calc_mre_K: 0.4515 - val_loss: 76.3105 - val_calc_mre_K: 0.4664\n", "Epoch 977/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 75.4693 - calc_mre_K: 0.4606 - val_loss: 76.9370 - val_calc_mre_K: 0.4702\n", "Epoch 978/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.5965 - calc_mre_K: 0.4553 - val_loss: 73.0665 - val_calc_mre_K: 0.4466\n", "Epoch 979/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.5811 - calc_mre_K: 0.4552 - val_loss: 72.6520 - val_calc_mre_K: 0.4440\n", "Epoch 980/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.9200 - calc_mre_K: 0.4573 - val_loss: 72.8998 - val_calc_mre_K: 0.4455\n", "Epoch 981/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.6380 - calc_mre_K: 0.4556 - val_loss: 70.9279 - val_calc_mre_K: 0.4335\n", "Epoch 982/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.9297 - calc_mre_K: 0.4512 - val_loss: 75.8468 - val_calc_mre_K: 0.4635\n", "Epoch 983/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 74.5800 - calc_mre_K: 0.4552 - val_loss: 81.6775 - val_calc_mre_K: 0.4992\n", "Epoch 984/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.6680 - calc_mre_K: 0.4557 - val_loss: 74.9662 - val_calc_mre_K: 0.4582\n", "Epoch 985/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.8676 - calc_mre_K: 0.4509 - val_loss: 74.6055 - val_calc_mre_K: 0.4559\n", "Epoch 986/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.5420 - calc_mre_K: 0.4550 - val_loss: 77.6354 - val_calc_mre_K: 0.4745\n", "Epoch 987/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.6315 - calc_mre_K: 0.4555 - val_loss: 73.4939 - val_calc_mre_K: 0.4492\n", "Epoch 988/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.0821 - calc_mre_K: 0.4522 - val_loss: 70.6545 - val_calc_mre_K: 0.4318\n", "Epoch 989/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.2878 - calc_mre_K: 0.4534 - val_loss: 74.6404 - val_calc_mre_K: 0.4562\n", "Epoch 990/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.4682 - calc_mre_K: 0.4545 - val_loss: 74.4092 - val_calc_mre_K: 0.4548\n", "Epoch 991/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.9641 - calc_mre_K: 0.4514 - val_loss: 72.6843 - val_calc_mre_K: 0.4442\n", "Epoch 992/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.9769 - calc_mre_K: 0.4576 - val_loss: 71.1184 - val_calc_mre_K: 0.4346\n", "Epoch 993/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.4580 - calc_mre_K: 0.4545 - val_loss: 72.8052 - val_calc_mre_K: 0.4449\n", "Epoch 994/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 74.3136 - calc_mre_K: 0.4536 - val_loss: 74.0963 - val_calc_mre_K: 0.4528\n", "Epoch 995/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.1102 - calc_mre_K: 0.4523 - val_loss: 71.3233 - val_calc_mre_K: 0.4359\n", "Epoch 996/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.0475 - calc_mre_K: 0.4520 - val_loss: 72.7472 - val_calc_mre_K: 0.4446\n", "Epoch 997/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.6638 - calc_mre_K: 0.4557 - val_loss: 72.8169 - val_calc_mre_K: 0.4450\n", "Epoch 998/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 74.0012 - calc_mre_K: 0.4517 - val_loss: 72.1107 - val_calc_mre_K: 0.4407\n", "Epoch 999/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.4687 - calc_mre_K: 0.4545 - val_loss: 76.5720 - val_calc_mre_K: 0.4680\n", "Epoch 1000/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 74.3055 - calc_mre_K: 0.4535 - val_loss: 71.9251 - val_calc_mre_K: 0.4396\n", "Epoch 1001/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.8572 - calc_mre_K: 0.4508 - val_loss: 76.8303 - val_calc_mre_K: 0.4695\n", "Epoch 1002/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.3850 - calc_mre_K: 0.4540 - val_loss: 75.3033 - val_calc_mre_K: 0.4602\n", "Epoch 1003/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.6486 - calc_mre_K: 0.4495 - val_loss: 73.8878 - val_calc_mre_K: 0.4516\n", "Epoch 1004/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 74.3100 - calc_mre_K: 0.4536 - val_loss: 75.5419 - val_calc_mre_K: 0.4617\n", "Epoch 1005/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 74.2662 - calc_mre_K: 0.4533 - val_loss: 75.8328 - val_calc_mre_K: 0.4634\n", "Epoch 1006/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.8503 - calc_mre_K: 0.4507 - val_loss: 71.3507 - val_calc_mre_K: 0.4361\n", "Epoch 1007/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.7503 - calc_mre_K: 0.4501 - val_loss: 71.3830 - val_calc_mre_K: 0.4363\n", "Epoch 1008/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.4632 - calc_mre_K: 0.4484 - val_loss: 72.3648 - val_calc_mre_K: 0.4423\n", "Epoch 1009/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 74.3063 - calc_mre_K: 0.4535 - val_loss: 72.7600 - val_calc_mre_K: 0.4447\n", "Epoch 1010/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 74.0049 - calc_mre_K: 0.4517 - val_loss: 75.0158 - val_calc_mre_K: 0.4584\n", "Epoch 1011/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.7361 - calc_mre_K: 0.4500 - val_loss: 74.2771 - val_calc_mre_K: 0.4539\n", "Epoch 1012/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 73.9049 - calc_mre_K: 0.4511 - val_loss: 74.5121 - val_calc_mre_K: 0.4554\n", "Epoch 1013/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 73.8674 - calc_mre_K: 0.4509 - val_loss: 73.8075 - val_calc_mre_K: 0.4511\n", "Epoch 1014/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.7948 - calc_mre_K: 0.4504 - val_loss: 74.5135 - val_calc_mre_K: 0.4554\n", "Epoch 1015/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 74.3711 - calc_mre_K: 0.4539 - val_loss: 73.7778 - val_calc_mre_K: 0.4509\n", "Epoch 1016/2000\n", "48000/48000 [==============================] - 3s 63us/step - loss: 73.7613 - calc_mre_K: 0.4502 - val_loss: 74.4738 - val_calc_mre_K: 0.4552\n", "Epoch 1017/2000\n", "48000/48000 [==============================] - 3s 65us/step - loss: 73.3118 - calc_mre_K: 0.4475 - val_loss: 71.8861 - val_calc_mre_K: 0.4393\n", "Epoch 1018/2000\n", "48000/48000 [==============================] - 3s 63us/step - loss: 74.4066 - calc_mre_K: 0.4541 - val_loss: 72.4393 - val_calc_mre_K: 0.4427\n", "Epoch 1019/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.9016 - calc_mre_K: 0.4511 - val_loss: 73.4148 - val_calc_mre_K: 0.4487\n", "Epoch 1020/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.4426 - calc_mre_K: 0.4483 - val_loss: 71.5158 - val_calc_mre_K: 0.4371\n", "Epoch 1021/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.8880 - calc_mre_K: 0.4510 - val_loss: 76.0612 - val_calc_mre_K: 0.4648\n", "Epoch 1022/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.5561 - calc_mre_K: 0.4490 - val_loss: 71.3459 - val_calc_mre_K: 0.4360\n", "Epoch 1023/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 74.2403 - calc_mre_K: 0.4531 - val_loss: 73.9561 - val_calc_mre_K: 0.4520\n", "Epoch 1024/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.8172 - calc_mre_K: 0.4505 - val_loss: 73.0344 - val_calc_mre_K: 0.4463\n", "Epoch 1025/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.2238 - calc_mre_K: 0.4469 - val_loss: 71.9480 - val_calc_mre_K: 0.4397\n", "Epoch 1026/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.3109 - calc_mre_K: 0.4475 - val_loss: 71.2607 - val_calc_mre_K: 0.4355\n", "Epoch 1027/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.7263 - calc_mre_K: 0.4500 - val_loss: 70.5154 - val_calc_mre_K: 0.4310\n", "Epoch 1028/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 73.5315 - calc_mre_K: 0.4488 - val_loss: 71.1776 - val_calc_mre_K: 0.4350\n", "Epoch 1029/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.3802 - calc_mre_K: 0.4479 - val_loss: 73.3091 - val_calc_mre_K: 0.4480\n", "Epoch 1030/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.6169 - calc_mre_K: 0.4493 - val_loss: 75.5876 - val_calc_mre_K: 0.4619\n", "Epoch 1031/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.6975 - calc_mre_K: 0.4498 - val_loss: 71.0834 - val_calc_mre_K: 0.4344\n", "Epoch 1032/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.4974 - calc_mre_K: 0.4486 - val_loss: 71.5740 - val_calc_mre_K: 0.4374\n", "Epoch 1033/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.4833 - calc_mre_K: 0.4485 - val_loss: 73.8183 - val_calc_mre_K: 0.4511\n", "Epoch 1034/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.6751 - calc_mre_K: 0.4497 - val_loss: 73.1682 - val_calc_mre_K: 0.4472\n", "Epoch 1035/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.3970 - calc_mre_K: 0.4480 - val_loss: 71.6091 - val_calc_mre_K: 0.4377\n", "Epoch 1036/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.7692 - calc_mre_K: 0.4503 - val_loss: 75.3054 - val_calc_mre_K: 0.4602\n", "Epoch 1037/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.1438 - calc_mre_K: 0.4464 - val_loss: 73.8511 - val_calc_mre_K: 0.4513\n", "Epoch 1038/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.4403 - calc_mre_K: 0.4482 - val_loss: 70.9854 - val_calc_mre_K: 0.4338\n", "Epoch 1039/2000\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "48000/48000 [==============================] - 3s 56us/step - loss: 72.7421 - calc_mre_K: 0.4440 - val_loss: 77.6350 - val_calc_mre_K: 0.4745\n", "Epoch 1040/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 73.3676 - calc_mre_K: 0.4478 - val_loss: 73.7999 - val_calc_mre_K: 0.4510\n", "Epoch 1041/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 73.9401 - calc_mre_K: 0.4513 - val_loss: 72.4538 - val_calc_mre_K: 0.4428\n", "Epoch 1042/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 73.8023 - calc_mre_K: 0.4505 - val_loss: 75.2394 - val_calc_mre_K: 0.4598\n", "Epoch 1043/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.3188 - calc_mre_K: 0.4475 - val_loss: 75.0123 - val_calc_mre_K: 0.4584\n", "Epoch 1044/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 73.0661 - calc_mre_K: 0.4460 - val_loss: 71.2796 - val_calc_mre_K: 0.4356\n", "Epoch 1045/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.4474 - calc_mre_K: 0.4422 - val_loss: 71.3523 - val_calc_mre_K: 0.4361\n", "Epoch 1046/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.5586 - calc_mre_K: 0.4490 - val_loss: 78.4876 - val_calc_mre_K: 0.4797\n", "Epoch 1047/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.3995 - calc_mre_K: 0.4480 - val_loss: 72.3766 - val_calc_mre_K: 0.4423\n", "Epoch 1048/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 72.7582 - calc_mre_K: 0.4441 - val_loss: 74.0821 - val_calc_mre_K: 0.4528\n", "Epoch 1049/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 73.1074 - calc_mre_K: 0.4462 - val_loss: 75.0532 - val_calc_mre_K: 0.4587\n", "Epoch 1050/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.4883 - calc_mre_K: 0.4424 - val_loss: 71.5145 - val_calc_mre_K: 0.4371\n", "Epoch 1051/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 73.2750 - calc_mre_K: 0.4472 - val_loss: 72.6006 - val_calc_mre_K: 0.4437\n", "Epoch 1052/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.6082 - calc_mre_K: 0.4493 - val_loss: 72.3976 - val_calc_mre_K: 0.4425\n", "Epoch 1053/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.7482 - calc_mre_K: 0.4440 - val_loss: 73.9845 - val_calc_mre_K: 0.4521\n", "Epoch 1054/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.4117 - calc_mre_K: 0.4481 - val_loss: 73.3677 - val_calc_mre_K: 0.4484\n", "Epoch 1055/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.7267 - calc_mre_K: 0.4439 - val_loss: 72.4663 - val_calc_mre_K: 0.4429\n", "Epoch 1056/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 73.0455 - calc_mre_K: 0.4458 - val_loss: 71.5749 - val_calc_mre_K: 0.4374\n", "Epoch 1057/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.8482 - calc_mre_K: 0.4446 - val_loss: 72.1739 - val_calc_mre_K: 0.4411\n", "Epoch 1058/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.1241 - calc_mre_K: 0.4463 - val_loss: 73.7847 - val_calc_mre_K: 0.4509\n", "Epoch 1059/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.3675 - calc_mre_K: 0.4478 - val_loss: 73.8078 - val_calc_mre_K: 0.4511\n", "Epoch 1060/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 73.3871 - calc_mre_K: 0.4479 - val_loss: 83.1494 - val_calc_mre_K: 0.5082\n", "Epoch 1061/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.6237 - calc_mre_K: 0.4433 - val_loss: 71.4621 - val_calc_mre_K: 0.4367\n", "Epoch 1062/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 72.4024 - calc_mre_K: 0.4419 - val_loss: 74.3322 - val_calc_mre_K: 0.4543\n", "Epoch 1063/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 73.2669 - calc_mre_K: 0.4472 - val_loss: 83.2605 - val_calc_mre_K: 0.5089\n", "Epoch 1064/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 73.9789 - calc_mre_K: 0.4515 - val_loss: 74.0143 - val_calc_mre_K: 0.4523\n", "Epoch 1065/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.5186 - calc_mre_K: 0.4426 - val_loss: 69.2546 - val_calc_mre_K: 0.4232\n", "Epoch 1066/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.5102 - calc_mre_K: 0.4426 - val_loss: 72.2202 - val_calc_mre_K: 0.4414\n", "Epoch 1067/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 72.6772 - calc_mre_K: 0.4436 - val_loss: 76.8868 - val_calc_mre_K: 0.4699\n", "Epoch 1068/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 73.6873 - calc_mre_K: 0.4498 - val_loss: 72.5725 - val_calc_mre_K: 0.4435\n", "Epoch 1069/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.0460 - calc_mre_K: 0.4397 - val_loss: 73.5921 - val_calc_mre_K: 0.4498\n", "Epoch 1070/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 72.5074 - calc_mre_K: 0.4425 - val_loss: 73.1841 - val_calc_mre_K: 0.4473\n", "Epoch 1071/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.6449 - calc_mre_K: 0.4434 - val_loss: 74.1972 - val_calc_mre_K: 0.4534\n", "Epoch 1072/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.5550 - calc_mre_K: 0.4428 - val_loss: 71.8098 - val_calc_mre_K: 0.4389\n", "Epoch 1073/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.6791 - calc_mre_K: 0.4436 - val_loss: 71.1314 - val_calc_mre_K: 0.4347\n", "Epoch 1074/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 73.4505 - calc_mre_K: 0.4483 - val_loss: 70.9270 - val_calc_mre_K: 0.4335\n", "Epoch 1075/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.4621 - calc_mre_K: 0.4423 - val_loss: 72.4792 - val_calc_mre_K: 0.4429\n", "Epoch 1076/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 72.3738 - calc_mre_K: 0.4417 - val_loss: 73.4052 - val_calc_mre_K: 0.4486\n", "Epoch 1077/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.1575 - calc_mre_K: 0.4465 - val_loss: 71.7944 - val_calc_mre_K: 0.4388\n", "Epoch 1078/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.6943 - calc_mre_K: 0.4437 - val_loss: 70.1882 - val_calc_mre_K: 0.4290\n", "Epoch 1079/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.3077 - calc_mre_K: 0.4413 - val_loss: 73.2773 - val_calc_mre_K: 0.4478\n", "Epoch 1080/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.8195 - calc_mre_K: 0.4445 - val_loss: 70.6751 - val_calc_mre_K: 0.4319\n", "Epoch 1081/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.7764 - calc_mre_K: 0.4442 - val_loss: 72.7359 - val_calc_mre_K: 0.4445\n", "Epoch 1082/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.6106 - calc_mre_K: 0.4432 - val_loss: 70.0374 - val_calc_mre_K: 0.4280\n", "Epoch 1083/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.1247 - calc_mre_K: 0.4463 - val_loss: 71.1388 - val_calc_mre_K: 0.4348\n", "Epoch 1084/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.6026 - calc_mre_K: 0.4431 - val_loss: 75.3451 - val_calc_mre_K: 0.4605\n", "Epoch 1085/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.6010 - calc_mre_K: 0.4431 - val_loss: 86.3586 - val_calc_mre_K: 0.5278\n", "Epoch 1086/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 72.4945 - calc_mre_K: 0.4425 - val_loss: 72.8210 - val_calc_mre_K: 0.4450\n", "Epoch 1087/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.6619 - calc_mre_K: 0.4435 - val_loss: 76.8988 - val_calc_mre_K: 0.4700\n", "Epoch 1088/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.1809 - calc_mre_K: 0.4467 - val_loss: 70.6589 - val_calc_mre_K: 0.4318\n", "Epoch 1089/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3655 - calc_mre_K: 0.4356 - val_loss: 76.1536 - val_calc_mre_K: 0.4654\n", "Epoch 1090/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 72.8035 - calc_mre_K: 0.4444 - val_loss: 76.3989 - val_calc_mre_K: 0.4669\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1091/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.9361 - calc_mre_K: 0.4452 - val_loss: 70.0537 - val_calc_mre_K: 0.4281\n", "Epoch 1092/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 72.0126 - calc_mre_K: 0.4395 - val_loss: 71.7583 - val_calc_mre_K: 0.4385\n", "Epoch 1093/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.6551 - calc_mre_K: 0.4435 - val_loss: 72.0408 - val_calc_mre_K: 0.4403\n", "Epoch 1094/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.3395 - calc_mre_K: 0.4415 - val_loss: 72.0190 - val_calc_mre_K: 0.4401\n", "Epoch 1095/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 73.2035 - calc_mre_K: 0.4468 - val_loss: 75.2397 - val_calc_mre_K: 0.4598\n", "Epoch 1096/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.1818 - calc_mre_K: 0.4406 - val_loss: 72.1392 - val_calc_mre_K: 0.4409\n", "Epoch 1097/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 73.2196 - calc_mre_K: 0.4469 - val_loss: 76.4734 - val_calc_mre_K: 0.4674\n", "Epoch 1098/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.9365 - calc_mre_K: 0.4391 - val_loss: 74.5687 - val_calc_mre_K: 0.4557\n", "Epoch 1099/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.4903 - calc_mre_K: 0.4424 - val_loss: 71.1336 - val_calc_mre_K: 0.4347\n", "Epoch 1100/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.6862 - calc_mre_K: 0.4436 - val_loss: 70.6135 - val_calc_mre_K: 0.4315\n", "Epoch 1101/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.9151 - calc_mre_K: 0.4389 - val_loss: 70.7169 - val_calc_mre_K: 0.4322\n", "Epoch 1102/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.5469 - calc_mre_K: 0.4428 - val_loss: 73.0073 - val_calc_mre_K: 0.4462\n", "Epoch 1103/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.9116 - calc_mre_K: 0.4389 - val_loss: 73.7251 - val_calc_mre_K: 0.4506\n", "Epoch 1104/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 73.8152 - calc_mre_K: 0.4505 - val_loss: 84.6053 - val_calc_mre_K: 0.5171\n", "Epoch 1105/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.6652 - calc_mre_K: 0.4374 - val_loss: 73.5211 - val_calc_mre_K: 0.4493\n", "Epoch 1106/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.1395 - calc_mre_K: 0.4403 - val_loss: 71.0152 - val_calc_mre_K: 0.4340\n", "Epoch 1107/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.0118 - calc_mre_K: 0.4395 - val_loss: 71.8143 - val_calc_mre_K: 0.4389\n", "Epoch 1108/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.6281 - calc_mre_K: 0.4433 - val_loss: 70.9508 - val_calc_mre_K: 0.4336\n", "Epoch 1109/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.3038 - calc_mre_K: 0.4413 - val_loss: 78.5762 - val_calc_mre_K: 0.4802\n", "Epoch 1110/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.6180 - calc_mre_K: 0.4371 - val_loss: 69.1685 - val_calc_mre_K: 0.4227\n", "Epoch 1111/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.5446 - calc_mre_K: 0.4428 - val_loss: 73.3988 - val_calc_mre_K: 0.4486\n", "Epoch 1112/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.9420 - calc_mre_K: 0.4391 - val_loss: 73.8309 - val_calc_mre_K: 0.4512\n", "Epoch 1113/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.3155 - calc_mre_K: 0.4414 - val_loss: 77.7378 - val_calc_mre_K: 0.4751\n", "Epoch 1114/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.9515 - calc_mre_K: 0.4392 - val_loss: 72.5649 - val_calc_mre_K: 0.4435\n", "Epoch 1115/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.1086 - calc_mre_K: 0.4401 - val_loss: 72.3016 - val_calc_mre_K: 0.4419\n", "Epoch 1116/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.0978 - calc_mre_K: 0.4400 - val_loss: 70.1451 - val_calc_mre_K: 0.4287\n", "Epoch 1117/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.5895 - calc_mre_K: 0.4431 - val_loss: 73.2292 - val_calc_mre_K: 0.4475\n", "Epoch 1118/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.8736 - calc_mre_K: 0.4448 - val_loss: 72.5284 - val_calc_mre_K: 0.4433\n", "Epoch 1119/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.6710 - calc_mre_K: 0.4374 - val_loss: 72.3817 - val_calc_mre_K: 0.4424\n", "Epoch 1120/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.9356 - calc_mre_K: 0.4391 - val_loss: 73.1365 - val_calc_mre_K: 0.4470\n", "Epoch 1121/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 72.2210 - calc_mre_K: 0.4408 - val_loss: 71.2982 - val_calc_mre_K: 0.4357\n", "Epoch 1122/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.7973 - calc_mre_K: 0.4382 - val_loss: 68.7672 - val_calc_mre_K: 0.4203\n", "Epoch 1123/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.6622 - calc_mre_K: 0.4374 - val_loss: 71.3266 - val_calc_mre_K: 0.4359\n", "Epoch 1124/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.0442 - calc_mre_K: 0.4397 - val_loss: 74.4640 - val_calc_mre_K: 0.4551\n", "Epoch 1125/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.4647 - calc_mre_K: 0.4423 - val_loss: 77.0824 - val_calc_mre_K: 0.4711\n", "Epoch 1126/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.7983 - calc_mre_K: 0.4382 - val_loss: 70.0904 - val_calc_mre_K: 0.4284\n", "Epoch 1127/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 72.1286 - calc_mre_K: 0.4402 - val_loss: 74.9010 - val_calc_mre_K: 0.4577\n", "Epoch 1128/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.0979 - calc_mre_K: 0.4401 - val_loss: 74.5947 - val_calc_mre_K: 0.4559\n", "Epoch 1129/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.1885 - calc_mre_K: 0.4406 - val_loss: 72.8925 - val_calc_mre_K: 0.4455\n", "Epoch 1130/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.7521 - calc_mre_K: 0.4379 - val_loss: 69.0226 - val_calc_mre_K: 0.4218\n", "Epoch 1131/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.0059 - calc_mre_K: 0.4395 - val_loss: 73.2372 - val_calc_mre_K: 0.4476\n", "Epoch 1132/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.2220 - calc_mre_K: 0.4408 - val_loss: 69.3686 - val_calc_mre_K: 0.4239\n", "Epoch 1133/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 72.3043 - calc_mre_K: 0.4413 - val_loss: 75.2404 - val_calc_mre_K: 0.4598\n", "Epoch 1134/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.9826 - calc_mre_K: 0.4332 - val_loss: 70.9837 - val_calc_mre_K: 0.4338\n", "Epoch 1135/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 71.8365 - calc_mre_K: 0.4385 - val_loss: 69.2073 - val_calc_mre_K: 0.4230\n", "Epoch 1136/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.8398 - calc_mre_K: 0.4385 - val_loss: 74.0443 - val_calc_mre_K: 0.4525\n", "Epoch 1137/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.4477 - calc_mre_K: 0.4361 - val_loss: 70.0869 - val_calc_mre_K: 0.4283\n", "Epoch 1138/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.2976 - calc_mre_K: 0.4413 - val_loss: 73.8456 - val_calc_mre_K: 0.4513\n", "Epoch 1139/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.1755 - calc_mre_K: 0.4405 - val_loss: 73.7363 - val_calc_mre_K: 0.4506\n", "Epoch 1140/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 72.4763 - calc_mre_K: 0.4424 - val_loss: 74.0514 - val_calc_mre_K: 0.4526\n", "Epoch 1141/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.5844 - calc_mre_K: 0.4369 - val_loss: 73.6122 - val_calc_mre_K: 0.4499\n", "Epoch 1142/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.6016 - calc_mre_K: 0.4370 - val_loss: 70.6422 - val_calc_mre_K: 0.4317\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1143/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 71.7588 - calc_mre_K: 0.4380 - val_loss: 69.8626 - val_calc_mre_K: 0.4270\n", "Epoch 1144/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.6038 - calc_mre_K: 0.4370 - val_loss: 71.8483 - val_calc_mre_K: 0.4391\n", "Epoch 1145/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 72.3257 - calc_mre_K: 0.4414 - val_loss: 72.0287 - val_calc_mre_K: 0.4402\n", "Epoch 1146/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.6514 - calc_mre_K: 0.4373 - val_loss: 76.0672 - val_calc_mre_K: 0.4649\n", "Epoch 1147/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.7718 - calc_mre_K: 0.4381 - val_loss: 69.7897 - val_calc_mre_K: 0.4265\n", "Epoch 1148/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 71.1029 - calc_mre_K: 0.4340 - val_loss: 69.7249 - val_calc_mre_K: 0.4261\n", "Epoch 1149/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 72.8948 - calc_mre_K: 0.4449 - val_loss: 73.7400 - val_calc_mre_K: 0.4507\n", "Epoch 1150/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 72.4694 - calc_mre_K: 0.4423 - val_loss: 69.8592 - val_calc_mre_K: 0.4269\n", "Epoch 1151/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 72.6466 - calc_mre_K: 0.4434 - val_loss: 79.6062 - val_calc_mre_K: 0.4865\n", "Epoch 1152/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.9083 - calc_mre_K: 0.4389 - val_loss: 69.9331 - val_calc_mre_K: 0.4274\n", "Epoch 1153/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.5348 - calc_mre_K: 0.4366 - val_loss: 68.1313 - val_calc_mre_K: 0.4164\n", "Epoch 1154/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.9212 - calc_mre_K: 0.4390 - val_loss: 70.9653 - val_calc_mre_K: 0.4337\n", "Epoch 1155/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.6512 - calc_mre_K: 0.4373 - val_loss: 72.7812 - val_calc_mre_K: 0.4448\n", "Epoch 1156/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.9905 - calc_mre_K: 0.4394 - val_loss: 73.2540 - val_calc_mre_K: 0.4477\n", "Epoch 1157/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.4852 - calc_mre_K: 0.4363 - val_loss: 70.8134 - val_calc_mre_K: 0.4328\n", "Epoch 1158/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.6166 - calc_mre_K: 0.4371 - val_loss: 69.7978 - val_calc_mre_K: 0.4266\n", "Epoch 1159/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.9668 - calc_mre_K: 0.4393 - val_loss: 71.5148 - val_calc_mre_K: 0.4371\n", "Epoch 1160/2000\n", "48000/48000 [==============================] - 3s 65us/step - loss: 71.4933 - calc_mre_K: 0.4364 - val_loss: 72.1562 - val_calc_mre_K: 0.4410\n", "Epoch 1161/2000\n", "48000/48000 [==============================] - 3s 66us/step - loss: 71.2561 - calc_mre_K: 0.4349 - val_loss: 73.0007 - val_calc_mre_K: 0.4461\n", "Epoch 1162/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.6304 - calc_mre_K: 0.4372 - val_loss: 72.4377 - val_calc_mre_K: 0.4427\n", "Epoch 1163/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.7238 - calc_mre_K: 0.4378 - val_loss: 79.0252 - val_calc_mre_K: 0.4829\n", "Epoch 1164/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.3855 - calc_mre_K: 0.4357 - val_loss: 71.1880 - val_calc_mre_K: 0.4351\n", "Epoch 1165/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.9710 - calc_mre_K: 0.4393 - val_loss: 72.2056 - val_calc_mre_K: 0.4413\n", "Epoch 1166/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.3027 - calc_mre_K: 0.4352 - val_loss: 74.3486 - val_calc_mre_K: 0.4544\n", "Epoch 1167/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.2215 - calc_mre_K: 0.4347 - val_loss: 70.2068 - val_calc_mre_K: 0.4291\n", "Epoch 1168/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3470 - calc_mre_K: 0.4355 - val_loss: 83.0142 - val_calc_mre_K: 0.5074\n", "Epoch 1169/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 71.7645 - calc_mre_K: 0.4380 - val_loss: 67.4809 - val_calc_mre_K: 0.4124\n", "Epoch 1170/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.2947 - calc_mre_K: 0.4351 - val_loss: 70.7765 - val_calc_mre_K: 0.4325\n", "Epoch 1171/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 72.0045 - calc_mre_K: 0.4395 - val_loss: 70.4802 - val_calc_mre_K: 0.4307\n", "Epoch 1172/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.1829 - calc_mre_K: 0.4345 - val_loss: 70.7713 - val_calc_mre_K: 0.4325\n", "Epoch 1173/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 71.3781 - calc_mre_K: 0.4357 - val_loss: 70.7346 - val_calc_mre_K: 0.4323\n", "Epoch 1174/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.5756 - calc_mre_K: 0.4369 - val_loss: 73.2221 - val_calc_mre_K: 0.4475\n", "Epoch 1175/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3179 - calc_mre_K: 0.4353 - val_loss: 79.3663 - val_calc_mre_K: 0.4850\n", "Epoch 1176/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3722 - calc_mre_K: 0.4356 - val_loss: 70.6470 - val_calc_mre_K: 0.4318\n", "Epoch 1177/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.9613 - calc_mre_K: 0.4331 - val_loss: 71.4322 - val_calc_mre_K: 0.4366\n", "Epoch 1178/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3915 - calc_mre_K: 0.4357 - val_loss: 69.9830 - val_calc_mre_K: 0.4277\n", "Epoch 1179/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.4988 - calc_mre_K: 0.4364 - val_loss: 72.1139 - val_calc_mre_K: 0.4407\n", "Epoch 1180/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 71.1277 - calc_mre_K: 0.4341 - val_loss: 71.7437 - val_calc_mre_K: 0.4385\n", "Epoch 1181/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.9915 - calc_mre_K: 0.4333 - val_loss: 77.0532 - val_calc_mre_K: 0.4709\n", "Epoch 1182/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.6645 - calc_mre_K: 0.4374 - val_loss: 68.1486 - val_calc_mre_K: 0.4165\n", "Epoch 1183/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.8972 - calc_mre_K: 0.4388 - val_loss: 73.8476 - val_calc_mre_K: 0.4513\n", "Epoch 1184/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.0880 - calc_mre_K: 0.4339 - val_loss: 71.9200 - val_calc_mre_K: 0.4395\n", "Epoch 1185/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.0049 - calc_mre_K: 0.4334 - val_loss: 71.2438 - val_calc_mre_K: 0.4354\n", "Epoch 1186/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.1325 - calc_mre_K: 0.4342 - val_loss: 79.0645 - val_calc_mre_K: 0.4832\n", "Epoch 1187/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 72.1993 - calc_mre_K: 0.4407 - val_loss: 68.9314 - val_calc_mre_K: 0.4213\n", "Epoch 1188/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3390 - calc_mre_K: 0.4354 - val_loss: 69.5201 - val_calc_mre_K: 0.4249\n", "Epoch 1189/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.5063 - calc_mre_K: 0.4364 - val_loss: 71.2404 - val_calc_mre_K: 0.4354\n", "Epoch 1190/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.3889 - calc_mre_K: 0.4357 - val_loss: 69.5828 - val_calc_mre_K: 0.4252\n", "Epoch 1191/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.0065 - calc_mre_K: 0.4334 - val_loss: 70.9738 - val_calc_mre_K: 0.4338\n", "Epoch 1192/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 71.0645 - calc_mre_K: 0.4337 - val_loss: 70.9960 - val_calc_mre_K: 0.4339\n", "Epoch 1193/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.0925 - calc_mre_K: 0.4339 - val_loss: 71.0852 - val_calc_mre_K: 0.4344\n", "Epoch 1194/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.6007 - calc_mre_K: 0.4370 - val_loss: 72.0862 - val_calc_mre_K: 0.4406\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1195/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.7482 - calc_mre_K: 0.4318 - val_loss: 72.1709 - val_calc_mre_K: 0.4411\n", "Epoch 1196/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.3808 - calc_mre_K: 0.4357 - val_loss: 69.0024 - val_calc_mre_K: 0.4217\n", "Epoch 1197/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.0520 - calc_mre_K: 0.4337 - val_loss: 81.9505 - val_calc_mre_K: 0.5008\n", "Epoch 1198/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.9719 - calc_mre_K: 0.4393 - val_loss: 71.6844 - val_calc_mre_K: 0.4381\n", "Epoch 1199/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 71.0513 - calc_mre_K: 0.4337 - val_loss: 72.7855 - val_calc_mre_K: 0.4448\n", "Epoch 1200/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.2043 - calc_mre_K: 0.4346 - val_loss: 71.8019 - val_calc_mre_K: 0.4388\n", "Epoch 1201/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 71.2594 - calc_mre_K: 0.4349 - val_loss: 70.7302 - val_calc_mre_K: 0.4323\n", "Epoch 1202/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3902 - calc_mre_K: 0.4357 - val_loss: 69.6456 - val_calc_mre_K: 0.4256\n", "Epoch 1203/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.5055 - calc_mre_K: 0.4303 - val_loss: 68.8535 - val_calc_mre_K: 0.4208\n", "Epoch 1204/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.7688 - calc_mre_K: 0.4380 - val_loss: 71.8588 - val_calc_mre_K: 0.4392\n", "Epoch 1205/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.8744 - calc_mre_K: 0.4326 - val_loss: 72.2768 - val_calc_mre_K: 0.4417\n", "Epoch 1206/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.2371 - calc_mre_K: 0.4348 - val_loss: 71.6431 - val_calc_mre_K: 0.4378\n", "Epoch 1207/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.2414 - calc_mre_K: 0.4348 - val_loss: 79.0945 - val_calc_mre_K: 0.4834\n", "Epoch 1208/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.4543 - calc_mre_K: 0.4361 - val_loss: 67.5974 - val_calc_mre_K: 0.4131\n", "Epoch 1209/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 71.4603 - calc_mre_K: 0.4362 - val_loss: 71.9457 - val_calc_mre_K: 0.4397\n", "Epoch 1210/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.8488 - calc_mre_K: 0.4324 - val_loss: 70.7912 - val_calc_mre_K: 0.4326\n", "Epoch 1211/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.9717 - calc_mre_K: 0.4332 - val_loss: 68.7572 - val_calc_mre_K: 0.4202\n", "Epoch 1212/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.7178 - calc_mre_K: 0.4316 - val_loss: 72.2083 - val_calc_mre_K: 0.4413\n", "Epoch 1213/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.7541 - calc_mre_K: 0.4380 - val_loss: 69.0467 - val_calc_mre_K: 0.4220\n", "Epoch 1214/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.5077 - calc_mre_K: 0.4303 - val_loss: 74.0668 - val_calc_mre_K: 0.4526\n", "Epoch 1215/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.5342 - calc_mre_K: 0.4366 - val_loss: 68.9654 - val_calc_mre_K: 0.4215\n", "Epoch 1216/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.1051 - calc_mre_K: 0.4340 - val_loss: 70.5662 - val_calc_mre_K: 0.4312\n", "Epoch 1217/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.7104 - calc_mre_K: 0.4316 - val_loss: 71.8305 - val_calc_mre_K: 0.4390\n", "Epoch 1218/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.9439 - calc_mre_K: 0.4330 - val_loss: 70.9907 - val_calc_mre_K: 0.4339\n", "Epoch 1219/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.1710 - calc_mre_K: 0.4344 - val_loss: 71.2842 - val_calc_mre_K: 0.4356\n", "Epoch 1220/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.7478 - calc_mre_K: 0.4318 - val_loss: 67.8759 - val_calc_mre_K: 0.4148\n", "Epoch 1221/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.1190 - calc_mre_K: 0.4341 - val_loss: 73.8166 - val_calc_mre_K: 0.4511\n", "Epoch 1222/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.8501 - calc_mre_K: 0.4324 - val_loss: 70.1164 - val_calc_mre_K: 0.4285\n", "Epoch 1223/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 70.2155 - calc_mre_K: 0.4286 - val_loss: 70.2015 - val_calc_mre_K: 0.4290\n", "Epoch 1224/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 71.2511 - calc_mre_K: 0.4349 - val_loss: 70.1448 - val_calc_mre_K: 0.4287\n", "Epoch 1225/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.5088 - calc_mre_K: 0.4304 - val_loss: 75.4243 - val_calc_mre_K: 0.4609\n", "Epoch 1226/2000\n", "48000/48000 [==============================] - 2s 49us/step - loss: 70.7870 - calc_mre_K: 0.4320 - val_loss: 73.2372 - val_calc_mre_K: 0.4476\n", "Epoch 1227/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.9565 - calc_mre_K: 0.4331 - val_loss: 70.7770 - val_calc_mre_K: 0.4326\n", "Epoch 1228/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.3339 - calc_mre_K: 0.4293 - val_loss: 71.3105 - val_calc_mre_K: 0.4358\n", "Epoch 1229/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.1147 - calc_mre_K: 0.4340 - val_loss: 72.1569 - val_calc_mre_K: 0.4410\n", "Epoch 1230/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 70.2704 - calc_mre_K: 0.4289 - val_loss: 71.8859 - val_calc_mre_K: 0.4393\n", "Epoch 1231/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 71.5551 - calc_mre_K: 0.4367 - val_loss: 69.2918 - val_calc_mre_K: 0.4235\n", "Epoch 1232/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.6074 - calc_mre_K: 0.4310 - val_loss: 69.1327 - val_calc_mre_K: 0.4225\n", "Epoch 1233/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.1133 - calc_mre_K: 0.4279 - val_loss: 70.9676 - val_calc_mre_K: 0.4337\n", "Epoch 1234/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.0705 - calc_mre_K: 0.4338 - val_loss: 76.4535 - val_calc_mre_K: 0.4672\n", "Epoch 1235/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 70.8247 - calc_mre_K: 0.4323 - val_loss: 71.1220 - val_calc_mre_K: 0.4347\n", "Epoch 1236/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.4870 - calc_mre_K: 0.4302 - val_loss: 73.7355 - val_calc_mre_K: 0.4506\n", "Epoch 1237/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 70.4482 - calc_mre_K: 0.4300 - val_loss: 70.7324 - val_calc_mre_K: 0.4323\n", "Epoch 1238/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.1350 - calc_mre_K: 0.4342 - val_loss: 73.4265 - val_calc_mre_K: 0.4487\n", "Epoch 1239/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.4449 - calc_mre_K: 0.4300 - val_loss: 70.9676 - val_calc_mre_K: 0.4337\n", "Epoch 1240/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 70.8190 - calc_mre_K: 0.4322 - val_loss: 69.2326 - val_calc_mre_K: 0.4231\n", "Epoch 1241/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.4003 - calc_mre_K: 0.4358 - val_loss: 68.2258 - val_calc_mre_K: 0.4170\n", "Epoch 1242/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.5429 - calc_mre_K: 0.4306 - val_loss: 68.1351 - val_calc_mre_K: 0.4164\n", "Epoch 1243/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.4670 - calc_mre_K: 0.4301 - val_loss: 71.6074 - val_calc_mre_K: 0.4376\n", "Epoch 1244/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.8747 - calc_mre_K: 0.4326 - val_loss: 70.2478 - val_calc_mre_K: 0.4293\n", "Epoch 1245/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.4002 - calc_mre_K: 0.4297 - val_loss: 69.1016 - val_calc_mre_K: 0.4223\n", "Epoch 1246/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.5588 - calc_mre_K: 0.4307 - val_loss: 70.1210 - val_calc_mre_K: 0.4286\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1247/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 71.0308 - calc_mre_K: 0.4335 - val_loss: 70.4263 - val_calc_mre_K: 0.4304\n", "Epoch 1248/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.5298 - calc_mre_K: 0.4305 - val_loss: 69.0273 - val_calc_mre_K: 0.4219\n", "Epoch 1249/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.7385 - calc_mre_K: 0.4318 - val_loss: 72.6243 - val_calc_mre_K: 0.4438\n", "Epoch 1250/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.5457 - calc_mre_K: 0.4306 - val_loss: 72.8127 - val_calc_mre_K: 0.4450\n", "Epoch 1251/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.3736 - calc_mre_K: 0.4295 - val_loss: 71.0937 - val_calc_mre_K: 0.4345\n", "Epoch 1252/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.9644 - calc_mre_K: 0.4331 - val_loss: 69.4244 - val_calc_mre_K: 0.4243\n", "Epoch 1253/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.7779 - calc_mre_K: 0.4320 - val_loss: 81.7390 - val_calc_mre_K: 0.4995\n", "Epoch 1254/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 71.0762 - calc_mre_K: 0.4338 - val_loss: 68.4682 - val_calc_mre_K: 0.4184\n", "Epoch 1255/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.8362 - calc_mre_K: 0.4323 - val_loss: 73.5919 - val_calc_mre_K: 0.4497\n", "Epoch 1256/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.4836 - calc_mre_K: 0.4302 - val_loss: 69.7403 - val_calc_mre_K: 0.4262\n", "Epoch 1257/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.7970 - calc_mre_K: 0.4260 - val_loss: 73.6482 - val_calc_mre_K: 0.4501\n", "Epoch 1258/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.5237 - calc_mre_K: 0.4304 - val_loss: 70.1542 - val_calc_mre_K: 0.4287\n", "Epoch 1259/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.8559 - calc_mre_K: 0.4325 - val_loss: 70.1680 - val_calc_mre_K: 0.4288\n", "Epoch 1260/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.5222 - calc_mre_K: 0.4304 - val_loss: 71.5291 - val_calc_mre_K: 0.4371\n", "Epoch 1261/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.3154 - calc_mre_K: 0.4292 - val_loss: 69.5691 - val_calc_mre_K: 0.4252\n", "Epoch 1262/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.4350 - calc_mre_K: 0.4299 - val_loss: 68.6107 - val_calc_mre_K: 0.4193\n", "Epoch 1263/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.7071 - calc_mre_K: 0.4316 - val_loss: 74.3168 - val_calc_mre_K: 0.4542\n", "Epoch 1264/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.2376 - calc_mre_K: 0.4287 - val_loss: 73.1734 - val_calc_mre_K: 0.4472\n", "Epoch 1265/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.9714 - calc_mre_K: 0.4332 - val_loss: 73.3937 - val_calc_mre_K: 0.4485\n", "Epoch 1266/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.7436 - calc_mre_K: 0.4318 - val_loss: 74.6966 - val_calc_mre_K: 0.4565\n", "Epoch 1267/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.1731 - calc_mre_K: 0.4283 - val_loss: 71.0721 - val_calc_mre_K: 0.4343\n", "Epoch 1268/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.1986 - calc_mre_K: 0.4346 - val_loss: 68.8924 - val_calc_mre_K: 0.4210\n", "Epoch 1269/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 70.1985 - calc_mre_K: 0.4285 - val_loss: 69.8133 - val_calc_mre_K: 0.4267\n", "Epoch 1270/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.6865 - calc_mre_K: 0.4314 - val_loss: 68.5785 - val_calc_mre_K: 0.4191\n", "Epoch 1271/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.1233 - calc_mre_K: 0.4280 - val_loss: 81.5094 - val_calc_mre_K: 0.4981\n", "Epoch 1272/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.5517 - calc_mre_K: 0.4306 - val_loss: 69.6947 - val_calc_mre_K: 0.4259\n", "Epoch 1273/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.1740 - calc_mre_K: 0.4283 - val_loss: 65.9663 - val_calc_mre_K: 0.4031\n", "Epoch 1274/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.4968 - calc_mre_K: 0.4303 - val_loss: 71.4605 - val_calc_mre_K: 0.4367\n", "Epoch 1275/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 70.4113 - calc_mre_K: 0.4298 - val_loss: 67.4185 - val_calc_mre_K: 0.4120\n", "Epoch 1276/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.8494 - calc_mre_K: 0.4324 - val_loss: 69.0280 - val_calc_mre_K: 0.4219\n", "Epoch 1277/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.8334 - calc_mre_K: 0.4262 - val_loss: 68.2743 - val_calc_mre_K: 0.4173\n", "Epoch 1278/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.5523 - calc_mre_K: 0.4306 - val_loss: 71.3827 - val_calc_mre_K: 0.4363\n", "Epoch 1279/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.2986 - calc_mre_K: 0.4291 - val_loss: 69.3097 - val_calc_mre_K: 0.4236\n", "Epoch 1280/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 69.8169 - calc_mre_K: 0.4261 - val_loss: 71.9878 - val_calc_mre_K: 0.4400\n", "Epoch 1281/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.2745 - calc_mre_K: 0.4289 - val_loss: 69.5744 - val_calc_mre_K: 0.4252\n", "Epoch 1282/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.9159 - calc_mre_K: 0.4267 - val_loss: 68.2162 - val_calc_mre_K: 0.4169\n", "Epoch 1283/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.9920 - calc_mre_K: 0.4272 - val_loss: 69.1375 - val_calc_mre_K: 0.4225\n", "Epoch 1284/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 71.0554 - calc_mre_K: 0.4337 - val_loss: 68.9475 - val_calc_mre_K: 0.4214\n", "Epoch 1285/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.0464 - calc_mre_K: 0.4275 - val_loss: 74.9178 - val_calc_mre_K: 0.4579\n", "Epoch 1286/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.1654 - calc_mre_K: 0.4283 - val_loss: 72.2721 - val_calc_mre_K: 0.4417\n", "Epoch 1287/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.9727 - calc_mre_K: 0.4271 - val_loss: 69.3909 - val_calc_mre_K: 0.4241\n", "Epoch 1288/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.0466 - calc_mre_K: 0.4275 - val_loss: 72.7290 - val_calc_mre_K: 0.4445\n", "Epoch 1289/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 70.2402 - calc_mre_K: 0.4287 - val_loss: 69.3614 - val_calc_mre_K: 0.4239\n", "Epoch 1290/2000\n", "48000/48000 [==============================] - 3s 65us/step - loss: 70.6281 - calc_mre_K: 0.4311 - val_loss: 67.5598 - val_calc_mre_K: 0.4129\n", "Epoch 1291/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 70.4525 - calc_mre_K: 0.4300 - val_loss: 68.9736 - val_calc_mre_K: 0.4215\n", "Epoch 1292/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.0717 - calc_mre_K: 0.4277 - val_loss: 72.9386 - val_calc_mre_K: 0.4458\n", "Epoch 1293/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.2631 - calc_mre_K: 0.4289 - val_loss: 67.9321 - val_calc_mre_K: 0.4152\n", "Epoch 1294/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.7505 - calc_mre_K: 0.4257 - val_loss: 77.7080 - val_calc_mre_K: 0.4749\n", "Epoch 1295/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 71.0254 - calc_mre_K: 0.4335 - val_loss: 68.7413 - val_calc_mre_K: 0.4201\n", "Epoch 1296/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.8781 - calc_mre_K: 0.4265 - val_loss: 68.1015 - val_calc_mre_K: 0.4162\n", "Epoch 1297/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.2163 - calc_mre_K: 0.4286 - val_loss: 70.1454 - val_calc_mre_K: 0.4287\n", "Epoch 1298/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 69.8347 - calc_mre_K: 0.4262 - val_loss: 70.8665 - val_calc_mre_K: 0.4331\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1299/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 71.3794 - calc_mre_K: 0.4357 - val_loss: 77.0868 - val_calc_mre_K: 0.4711\n", "Epoch 1300/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.6787 - calc_mre_K: 0.4253 - val_loss: 69.4070 - val_calc_mre_K: 0.4242\n", "Epoch 1301/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 70.4443 - calc_mre_K: 0.4300 - val_loss: 70.1432 - val_calc_mre_K: 0.4287\n", "Epoch 1302/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 69.9951 - calc_mre_K: 0.4272 - val_loss: 68.7678 - val_calc_mre_K: 0.4203\n", "Epoch 1303/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.1724 - calc_mre_K: 0.4283 - val_loss: 71.0268 - val_calc_mre_K: 0.4341\n", "Epoch 1304/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.2740 - calc_mre_K: 0.4289 - val_loss: 76.4377 - val_calc_mre_K: 0.4672\n", "Epoch 1305/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.6840 - calc_mre_K: 0.4253 - val_loss: 69.1976 - val_calc_mre_K: 0.4229\n", "Epoch 1306/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.3362 - calc_mre_K: 0.4293 - val_loss: 70.1695 - val_calc_mre_K: 0.4288\n", "Epoch 1307/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 69.9956 - calc_mre_K: 0.4272 - val_loss: 71.6247 - val_calc_mre_K: 0.4377\n", "Epoch 1308/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.4603 - calc_mre_K: 0.4301 - val_loss: 67.9808 - val_calc_mre_K: 0.4155\n", "Epoch 1309/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.9420 - calc_mre_K: 0.4269 - val_loss: 70.4424 - val_calc_mre_K: 0.4305\n", "Epoch 1310/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.1833 - calc_mre_K: 0.4284 - val_loss: 70.2834 - val_calc_mre_K: 0.4295\n", "Epoch 1311/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.9718 - calc_mre_K: 0.4271 - val_loss: 67.3791 - val_calc_mre_K: 0.4118\n", "Epoch 1312/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 69.5862 - calc_mre_K: 0.4247 - val_loss: 74.9440 - val_calc_mre_K: 0.4580\n", "Epoch 1313/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.0547 - calc_mre_K: 0.4276 - val_loss: 72.7270 - val_calc_mre_K: 0.4445\n", "Epoch 1314/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.8352 - calc_mre_K: 0.4262 - val_loss: 69.4847 - val_calc_mre_K: 0.4247\n", "Epoch 1315/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 70.5802 - calc_mre_K: 0.4308 - val_loss: 68.4023 - val_calc_mre_K: 0.4180\n", "Epoch 1316/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.0768 - calc_mre_K: 0.4277 - val_loss: 70.3887 - val_calc_mre_K: 0.4302\n", "Epoch 1317/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.7890 - calc_mre_K: 0.4260 - val_loss: 70.5411 - val_calc_mre_K: 0.4311\n", "Epoch 1318/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.6047 - calc_mre_K: 0.4248 - val_loss: 67.7301 - val_calc_mre_K: 0.4139\n", "Epoch 1319/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.9106 - calc_mre_K: 0.4267 - val_loss: 69.4309 - val_calc_mre_K: 0.4243\n", "Epoch 1320/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.3480 - calc_mre_K: 0.4294 - val_loss: 68.3815 - val_calc_mre_K: 0.4179\n", "Epoch 1321/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 69.1704 - calc_mre_K: 0.4222 - val_loss: 69.3655 - val_calc_mre_K: 0.4239\n", "Epoch 1322/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 70.6502 - calc_mre_K: 0.4312 - val_loss: 71.6209 - val_calc_mre_K: 0.4377\n", "Epoch 1323/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 69.5715 - calc_mre_K: 0.4246 - val_loss: 72.2344 - val_calc_mre_K: 0.4415\n", "Epoch 1324/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.9795 - calc_mre_K: 0.4271 - val_loss: 76.9549 - val_calc_mre_K: 0.4703\n", "Epoch 1325/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.6182 - calc_mre_K: 0.4310 - val_loss: 72.5023 - val_calc_mre_K: 0.4431\n", "Epoch 1326/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.0654 - calc_mre_K: 0.4215 - val_loss: 69.5943 - val_calc_mre_K: 0.4253\n", "Epoch 1327/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.8812 - calc_mre_K: 0.4326 - val_loss: 72.5144 - val_calc_mre_K: 0.4432\n", "Epoch 1328/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 70.2488 - calc_mre_K: 0.4288 - val_loss: 68.2601 - val_calc_mre_K: 0.4172\n", "Epoch 1329/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 69.9083 - calc_mre_K: 0.4267 - val_loss: 69.6548 - val_calc_mre_K: 0.4257\n", "Epoch 1330/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.9074 - calc_mre_K: 0.4267 - val_loss: 72.2517 - val_calc_mre_K: 0.4416\n", "Epoch 1331/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 69.7868 - calc_mre_K: 0.4259 - val_loss: 71.0607 - val_calc_mre_K: 0.4343\n", "Epoch 1332/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.3428 - calc_mre_K: 0.4232 - val_loss: 76.5813 - val_calc_mre_K: 0.4680\n", "Epoch 1333/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.8964 - calc_mre_K: 0.4266 - val_loss: 67.6591 - val_calc_mre_K: 0.4135\n", "Epoch 1334/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.9852 - calc_mre_K: 0.4272 - val_loss: 67.4130 - val_calc_mre_K: 0.4120\n", "Epoch 1335/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.6813 - calc_mre_K: 0.4314 - val_loss: 73.2049 - val_calc_mre_K: 0.4474\n", "Epoch 1336/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.8770 - calc_mre_K: 0.4265 - val_loss: 68.8735 - val_calc_mre_K: 0.4209\n", "Epoch 1337/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.5391 - calc_mre_K: 0.4244 - val_loss: 72.1449 - val_calc_mre_K: 0.4409\n", "Epoch 1338/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.0340 - calc_mre_K: 0.4275 - val_loss: 66.8531 - val_calc_mre_K: 0.4086\n", "Epoch 1339/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 69.1922 - calc_mre_K: 0.4223 - val_loss: 75.8762 - val_calc_mre_K: 0.4637\n", "Epoch 1340/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.4689 - calc_mre_K: 0.4240 - val_loss: 68.6445 - val_calc_mre_K: 0.4195\n", "Epoch 1341/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.7994 - calc_mre_K: 0.4260 - val_loss: 70.0435 - val_calc_mre_K: 0.4281\n", "Epoch 1342/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.6903 - calc_mre_K: 0.4254 - val_loss: 68.0647 - val_calc_mre_K: 0.4160\n", "Epoch 1343/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.7890 - calc_mre_K: 0.4260 - val_loss: 67.6846 - val_calc_mre_K: 0.4137\n", "Epoch 1344/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 69.3899 - calc_mre_K: 0.4235 - val_loss: 68.6710 - val_calc_mre_K: 0.4197\n", "Epoch 1345/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 70.0653 - calc_mre_K: 0.4276 - val_loss: 68.5527 - val_calc_mre_K: 0.4190\n", "Epoch 1346/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.7421 - calc_mre_K: 0.4257 - val_loss: 70.9387 - val_calc_mre_K: 0.4335\n", "Epoch 1347/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 69.9003 - calc_mre_K: 0.4266 - val_loss: 75.6581 - val_calc_mre_K: 0.4624\n", "Epoch 1348/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.9036 - calc_mre_K: 0.4267 - val_loss: 72.3071 - val_calc_mre_K: 0.4419\n", "Epoch 1349/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.6799 - calc_mre_K: 0.4253 - val_loss: 72.6530 - val_calc_mre_K: 0.4440\n", "Epoch 1350/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.9104 - calc_mre_K: 0.4267 - val_loss: 68.3132 - val_calc_mre_K: 0.4175\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1351/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.2826 - calc_mre_K: 0.4229 - val_loss: 71.8987 - val_calc_mre_K: 0.4394\n", "Epoch 1352/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 69.5693 - calc_mre_K: 0.4246 - val_loss: 68.8476 - val_calc_mre_K: 0.4208\n", "Epoch 1353/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.5850 - calc_mre_K: 0.4247 - val_loss: 73.2841 - val_calc_mre_K: 0.4479\n", "Epoch 1354/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 70.1795 - calc_mre_K: 0.4283 - val_loss: 68.5144 - val_calc_mre_K: 0.4187\n", "Epoch 1355/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.7186 - calc_mre_K: 0.4255 - val_loss: 75.4468 - val_calc_mre_K: 0.4611\n", "Epoch 1356/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.5162 - calc_mre_K: 0.4243 - val_loss: 68.8912 - val_calc_mre_K: 0.4210\n", "Epoch 1357/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.4047 - calc_mre_K: 0.4236 - val_loss: 67.2678 - val_calc_mre_K: 0.4111\n", "Epoch 1358/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.4585 - calc_mre_K: 0.4239 - val_loss: 69.4207 - val_calc_mre_K: 0.4243\n", "Epoch 1359/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.7570 - calc_mre_K: 0.4258 - val_loss: 69.4304 - val_calc_mre_K: 0.4243\n", "Epoch 1360/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 69.4086 - calc_mre_K: 0.4236 - val_loss: 68.0590 - val_calc_mre_K: 0.4159\n", "Epoch 1361/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.8641 - calc_mre_K: 0.4264 - val_loss: 69.1673 - val_calc_mre_K: 0.4227\n", "Epoch 1362/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.2491 - calc_mre_K: 0.4227 - val_loss: 66.4902 - val_calc_mre_K: 0.4064\n", "Epoch 1363/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.2924 - calc_mre_K: 0.4290 - val_loss: 75.7687 - val_calc_mre_K: 0.4630\n", "Epoch 1364/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 69.3860 - calc_mre_K: 0.4235 - val_loss: 67.8990 - val_calc_mre_K: 0.4150\n", "Epoch 1365/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 69.2259 - calc_mre_K: 0.4225 - val_loss: 69.5974 - val_calc_mre_K: 0.4253\n", "Epoch 1366/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 69.9082 - calc_mre_K: 0.4267 - val_loss: 69.8654 - val_calc_mre_K: 0.4270\n", "Epoch 1367/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.9393 - calc_mre_K: 0.4208 - val_loss: 67.3726 - val_calc_mre_K: 0.4118\n", "Epoch 1368/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.3251 - calc_mre_K: 0.4231 - val_loss: 68.6787 - val_calc_mre_K: 0.4197\n", "Epoch 1369/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.1130 - calc_mre_K: 0.4218 - val_loss: 67.5085 - val_calc_mre_K: 0.4126\n", "Epoch 1370/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 70.1514 - calc_mre_K: 0.4282 - val_loss: 70.8700 - val_calc_mre_K: 0.4331\n", "Epoch 1371/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 69.6252 - calc_mre_K: 0.4250 - val_loss: 72.8357 - val_calc_mre_K: 0.4451\n", "Epoch 1372/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 69.2902 - calc_mre_K: 0.4229 - val_loss: 70.2126 - val_calc_mre_K: 0.4291\n", "Epoch 1373/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.3986 - calc_mre_K: 0.4236 - val_loss: 68.4818 - val_calc_mre_K: 0.4185\n", "Epoch 1374/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.0636 - calc_mre_K: 0.4215 - val_loss: 69.2609 - val_calc_mre_K: 0.4233\n", "Epoch 1375/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 70.9332 - calc_mre_K: 0.4329 - val_loss: 72.3035 - val_calc_mre_K: 0.4419\n", "Epoch 1376/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.4708 - calc_mre_K: 0.4240 - val_loss: 66.6590 - val_calc_mre_K: 0.4074\n", "Epoch 1377/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.2128 - calc_mre_K: 0.4224 - val_loss: 68.1624 - val_calc_mre_K: 0.4166\n", "Epoch 1378/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.4519 - calc_mre_K: 0.4239 - val_loss: 70.5474 - val_calc_mre_K: 0.4311\n", "Epoch 1379/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.1107 - calc_mre_K: 0.4218 - val_loss: 67.2729 - val_calc_mre_K: 0.4111\n", "Epoch 1380/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.0322 - calc_mre_K: 0.4213 - val_loss: 68.9509 - val_calc_mre_K: 0.4214\n", "Epoch 1381/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 69.7869 - calc_mre_K: 0.4259 - val_loss: 66.5613 - val_calc_mre_K: 0.4068\n", "Epoch 1382/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.0680 - calc_mre_K: 0.4216 - val_loss: 68.9872 - val_calc_mre_K: 0.4216\n", "Epoch 1383/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.5884 - calc_mre_K: 0.4186 - val_loss: 67.6452 - val_calc_mre_K: 0.4134\n", "Epoch 1384/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.2059 - calc_mre_K: 0.4224 - val_loss: 66.2243 - val_calc_mre_K: 0.4047\n", "Epoch 1385/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.1766 - calc_mre_K: 0.4222 - val_loss: 70.1962 - val_calc_mre_K: 0.4290\n", "Epoch 1386/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.4524 - calc_mre_K: 0.4239 - val_loss: 69.5569 - val_calc_mre_K: 0.4251\n", "Epoch 1387/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.6832 - calc_mre_K: 0.4253 - val_loss: 66.4420 - val_calc_mre_K: 0.4061\n", "Epoch 1388/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.4013 - calc_mre_K: 0.4236 - val_loss: 69.6874 - val_calc_mre_K: 0.4259\n", "Epoch 1389/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.7763 - calc_mre_K: 0.4198 - val_loss: 66.6848 - val_calc_mre_K: 0.4075\n", "Epoch 1390/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.1430 - calc_mre_K: 0.4220 - val_loss: 70.0602 - val_calc_mre_K: 0.4282\n", "Epoch 1391/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 70.1538 - calc_mre_K: 0.4282 - val_loss: 68.5372 - val_calc_mre_K: 0.4189\n", "Epoch 1392/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.4912 - calc_mre_K: 0.4180 - val_loss: 69.3859 - val_calc_mre_K: 0.4241\n", "Epoch 1393/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 69.3049 - calc_mre_K: 0.4230 - val_loss: 66.8168 - val_calc_mre_K: 0.4083\n", "Epoch 1394/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.6236 - calc_mre_K: 0.4249 - val_loss: 71.6569 - val_calc_mre_K: 0.4379\n", "Epoch 1395/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 68.8874 - calc_mre_K: 0.4205 - val_loss: 68.3673 - val_calc_mre_K: 0.4178\n", "Epoch 1396/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.7924 - calc_mre_K: 0.4199 - val_loss: 69.0296 - val_calc_mre_K: 0.4219\n", "Epoch 1397/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.6839 - calc_mre_K: 0.4253 - val_loss: 69.5192 - val_calc_mre_K: 0.4249\n", "Epoch 1398/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.5053 - calc_mre_K: 0.4242 - val_loss: 67.5787 - val_calc_mre_K: 0.4130\n", "Epoch 1399/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 69.5480 - calc_mre_K: 0.4245 - val_loss: 67.5165 - val_calc_mre_K: 0.4126\n", "Epoch 1400/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.1255 - calc_mre_K: 0.4219 - val_loss: 70.3795 - val_calc_mre_K: 0.4301\n", "Epoch 1401/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.9703 - calc_mre_K: 0.4210 - val_loss: 68.4392 - val_calc_mre_K: 0.4183\n", "Epoch 1402/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 68.8585 - calc_mre_K: 0.4203 - val_loss: 72.2691 - val_calc_mre_K: 0.4417\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1403/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.2529 - calc_mre_K: 0.4227 - val_loss: 67.0211 - val_calc_mre_K: 0.4096\n", "Epoch 1404/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.4711 - calc_mre_K: 0.4240 - val_loss: 68.7867 - val_calc_mre_K: 0.4204\n", "Epoch 1405/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 69.1807 - calc_mre_K: 0.4222 - val_loss: 68.2370 - val_calc_mre_K: 0.4170\n", "Epoch 1406/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 69.0415 - calc_mre_K: 0.4214 - val_loss: 72.4768 - val_calc_mre_K: 0.4429\n", "Epoch 1407/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.3495 - calc_mre_K: 0.4233 - val_loss: 71.0208 - val_calc_mre_K: 0.4341\n", "Epoch 1408/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.1211 - calc_mre_K: 0.4219 - val_loss: 68.1753 - val_calc_mre_K: 0.4167\n", "Epoch 1409/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.4316 - calc_mre_K: 0.4238 - val_loss: 70.4297 - val_calc_mre_K: 0.4304\n", "Epoch 1410/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.2327 - calc_mre_K: 0.4165 - val_loss: 71.6858 - val_calc_mre_K: 0.4381\n", "Epoch 1411/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.8780 - calc_mre_K: 0.4204 - val_loss: 70.7408 - val_calc_mre_K: 0.4323\n", "Epoch 1412/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.8486 - calc_mre_K: 0.4202 - val_loss: 72.4994 - val_calc_mre_K: 0.4431\n", "Epoch 1413/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.0058 - calc_mre_K: 0.4212 - val_loss: 68.7159 - val_calc_mre_K: 0.4199\n", "Epoch 1414/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 69.0153 - calc_mre_K: 0.4212 - val_loss: 75.3577 - val_calc_mre_K: 0.4606\n", "Epoch 1415/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.9862 - calc_mre_K: 0.4211 - val_loss: 68.1452 - val_calc_mre_K: 0.4165\n", "Epoch 1416/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.2629 - calc_mre_K: 0.4227 - val_loss: 69.8466 - val_calc_mre_K: 0.4269\n", "Epoch 1417/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.6826 - calc_mre_K: 0.4192 - val_loss: 72.5504 - val_calc_mre_K: 0.4434\n", "Epoch 1418/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.5717 - calc_mre_K: 0.4185 - val_loss: 66.5535 - val_calc_mre_K: 0.4067\n", "Epoch 1419/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.5399 - calc_mre_K: 0.4244 - val_loss: 68.4778 - val_calc_mre_K: 0.4185\n", "Epoch 1420/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 68.2100 - calc_mre_K: 0.4163 - val_loss: 70.8915 - val_calc_mre_K: 0.4332\n", "Epoch 1421/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.4497 - calc_mre_K: 0.4178 - val_loss: 66.8683 - val_calc_mre_K: 0.4087\n", "Epoch 1422/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.9545 - calc_mre_K: 0.4209 - val_loss: 71.0497 - val_calc_mre_K: 0.4342\n", "Epoch 1423/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.2151 - calc_mre_K: 0.4225 - val_loss: 70.8271 - val_calc_mre_K: 0.4329\n", "Epoch 1424/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.4184 - calc_mre_K: 0.4176 - val_loss: 68.6246 - val_calc_mre_K: 0.4194\n", "Epoch 1425/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 68.4284 - calc_mre_K: 0.4177 - val_loss: 69.4976 - val_calc_mre_K: 0.4247\n", "Epoch 1426/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 68.9536 - calc_mre_K: 0.4209 - val_loss: 67.9751 - val_calc_mre_K: 0.4154\n", "Epoch 1427/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.4917 - calc_mre_K: 0.4180 - val_loss: 69.7578 - val_calc_mre_K: 0.4263\n", "Epoch 1428/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 68.6341 - calc_mre_K: 0.4189 - val_loss: 70.0588 - val_calc_mre_K: 0.4282\n", "Epoch 1429/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.4955 - calc_mre_K: 0.4242 - val_loss: 66.8253 - val_calc_mre_K: 0.4084\n", "Epoch 1430/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.7581 - calc_mre_K: 0.4197 - val_loss: 66.7412 - val_calc_mre_K: 0.4079\n", "Epoch 1431/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 69.0456 - calc_mre_K: 0.4214 - val_loss: 66.9825 - val_calc_mre_K: 0.4094\n", "Epoch 1432/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.9314 - calc_mre_K: 0.4207 - val_loss: 69.6140 - val_calc_mre_K: 0.4254\n", "Epoch 1433/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.0745 - calc_mre_K: 0.4216 - val_loss: 65.9268 - val_calc_mre_K: 0.4029\n", "Epoch 1434/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 68.6886 - calc_mre_K: 0.4192 - val_loss: 71.0439 - val_calc_mre_K: 0.4342\n", "Epoch 1435/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.6053 - calc_mre_K: 0.4187 - val_loss: 70.9530 - val_calc_mre_K: 0.4336\n", "Epoch 1436/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.7735 - calc_mre_K: 0.4198 - val_loss: 68.5941 - val_calc_mre_K: 0.4192\n", "Epoch 1437/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.9954 - calc_mre_K: 0.4211 - val_loss: 70.4060 - val_calc_mre_K: 0.4303\n", "Epoch 1438/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.5182 - calc_mre_K: 0.4182 - val_loss: 67.1093 - val_calc_mre_K: 0.4101\n", "Epoch 1439/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.5459 - calc_mre_K: 0.4184 - val_loss: 65.4218 - val_calc_mre_K: 0.3998\n", "Epoch 1440/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.4335 - calc_mre_K: 0.4177 - val_loss: 69.4076 - val_calc_mre_K: 0.4242\n", "Epoch 1441/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.9414 - calc_mre_K: 0.4208 - val_loss: 85.4045 - val_calc_mre_K: 0.5219\n", "Epoch 1442/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.5285 - calc_mre_K: 0.4183 - val_loss: 68.3017 - val_calc_mre_K: 0.4174\n", "Epoch 1443/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 68.7941 - calc_mre_K: 0.4199 - val_loss: 70.5987 - val_calc_mre_K: 0.4315\n", "Epoch 1444/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.1992 - calc_mre_K: 0.4163 - val_loss: 67.6171 - val_calc_mre_K: 0.4132\n", "Epoch 1445/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.4160 - calc_mre_K: 0.4176 - val_loss: 69.1077 - val_calc_mre_K: 0.4224\n", "Epoch 1446/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.1675 - calc_mre_K: 0.4222 - val_loss: 66.8692 - val_calc_mre_K: 0.4087\n", "Epoch 1447/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.1412 - calc_mre_K: 0.4220 - val_loss: 68.5321 - val_calc_mre_K: 0.4188\n", "Epoch 1448/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.0518 - calc_mre_K: 0.4154 - val_loss: 70.3387 - val_calc_mre_K: 0.4299\n", "Epoch 1449/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.1951 - calc_mre_K: 0.4162 - val_loss: 69.8300 - val_calc_mre_K: 0.4268\n", "Epoch 1450/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 68.5539 - calc_mre_K: 0.4184 - val_loss: 68.5058 - val_calc_mre_K: 0.4187\n", "Epoch 1451/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.3328 - calc_mre_K: 0.4171 - val_loss: 70.1874 - val_calc_mre_K: 0.4290\n", "Epoch 1452/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.2736 - calc_mre_K: 0.4167 - val_loss: 75.9388 - val_calc_mre_K: 0.4641\n", "Epoch 1453/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 68.3581 - calc_mre_K: 0.4172 - val_loss: 66.8162 - val_calc_mre_K: 0.4084\n", "Epoch 1454/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.4491 - calc_mre_K: 0.4178 - val_loss: 71.2243 - val_calc_mre_K: 0.4353\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1455/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 68.1299 - calc_mre_K: 0.4158 - val_loss: 67.2592 - val_calc_mre_K: 0.4111\n", "Epoch 1456/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.9475 - calc_mre_K: 0.4208 - val_loss: 65.4615 - val_calc_mre_K: 0.4001\n", "Epoch 1457/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.5235 - calc_mre_K: 0.4182 - val_loss: 70.1915 - val_calc_mre_K: 0.4290\n", "Epoch 1458/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.1614 - calc_mre_K: 0.4160 - val_loss: 68.1668 - val_calc_mre_K: 0.4166\n", "Epoch 1459/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.3551 - calc_mre_K: 0.4172 - val_loss: 66.4457 - val_calc_mre_K: 0.4061\n", "Epoch 1460/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.3074 - calc_mre_K: 0.4169 - val_loss: 72.9080 - val_calc_mre_K: 0.4456\n", "Epoch 1461/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 68.3340 - calc_mre_K: 0.4171 - val_loss: 74.6191 - val_calc_mre_K: 0.4560\n", "Epoch 1462/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 68.4176 - calc_mre_K: 0.4176 - val_loss: 69.9668 - val_calc_mre_K: 0.4276\n", "Epoch 1463/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.9439 - calc_mre_K: 0.4147 - val_loss: 67.5213 - val_calc_mre_K: 0.4127\n", "Epoch 1464/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 69.1422 - calc_mre_K: 0.4220 - val_loss: 66.2677 - val_calc_mre_K: 0.4050\n", "Epoch 1465/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.9040 - calc_mre_K: 0.4145 - val_loss: 65.5460 - val_calc_mre_K: 0.4006\n", "Epoch 1466/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.4695 - calc_mre_K: 0.4179 - val_loss: 71.3172 - val_calc_mre_K: 0.4359\n", "Epoch 1467/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.2651 - calc_mre_K: 0.4167 - val_loss: 64.8670 - val_calc_mre_K: 0.3964\n", "Epoch 1468/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.8564 - calc_mre_K: 0.4142 - val_loss: 82.1148 - val_calc_mre_K: 0.5018\n", "Epoch 1469/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.7232 - calc_mre_K: 0.4195 - val_loss: 68.1878 - val_calc_mre_K: 0.4167\n", "Epoch 1470/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 68.2692 - calc_mre_K: 0.4167 - val_loss: 74.5817 - val_calc_mre_K: 0.4558\n", "Epoch 1471/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 68.0842 - calc_mre_K: 0.4156 - val_loss: 68.7380 - val_calc_mre_K: 0.4201\n", "Epoch 1472/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.5334 - calc_mre_K: 0.4183 - val_loss: 66.4642 - val_calc_mre_K: 0.4062\n", "Epoch 1473/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.3021 - calc_mre_K: 0.4169 - val_loss: 69.4432 - val_calc_mre_K: 0.4244\n", "Epoch 1474/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.2422 - calc_mre_K: 0.4165 - val_loss: 67.8014 - val_calc_mre_K: 0.4144\n", "Epoch 1475/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.8805 - calc_mre_K: 0.4143 - val_loss: 71.6950 - val_calc_mre_K: 0.4382\n", "Epoch 1476/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 69.0333 - calc_mre_K: 0.4213 - val_loss: 68.6462 - val_calc_mre_K: 0.4195\n", "Epoch 1477/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.1800 - calc_mre_K: 0.4161 - val_loss: 68.8780 - val_calc_mre_K: 0.4209\n", "Epoch 1478/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.7787 - calc_mre_K: 0.4137 - val_loss: 68.8928 - val_calc_mre_K: 0.4210\n", "Epoch 1479/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.8454 - calc_mre_K: 0.4141 - val_loss: 71.3600 - val_calc_mre_K: 0.4361\n", "Epoch 1480/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 69.1391 - calc_mre_K: 0.4220 - val_loss: 74.2008 - val_calc_mre_K: 0.4535\n", "Epoch 1481/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.7904 - calc_mre_K: 0.4138 - val_loss: 68.0249 - val_calc_mre_K: 0.4157\n", "Epoch 1482/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.9335 - calc_mre_K: 0.4146 - val_loss: 67.7388 - val_calc_mre_K: 0.4140\n", "Epoch 1483/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.0385 - calc_mre_K: 0.4153 - val_loss: 66.6939 - val_calc_mre_K: 0.4076\n", "Epoch 1484/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.3688 - calc_mre_K: 0.4173 - val_loss: 69.9214 - val_calc_mre_K: 0.4273\n", "Epoch 1485/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 68.8053 - calc_mre_K: 0.4200 - val_loss: 66.5163 - val_calc_mre_K: 0.4065\n", "Epoch 1486/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.5740 - calc_mre_K: 0.4124 - val_loss: 68.7154 - val_calc_mre_K: 0.4200\n", "Epoch 1487/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 67.7121 - calc_mre_K: 0.4133 - val_loss: 80.2742 - val_calc_mre_K: 0.4906\n", "Epoch 1488/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 68.3509 - calc_mre_K: 0.4172 - val_loss: 66.6684 - val_calc_mre_K: 0.4075\n", "Epoch 1489/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.8930 - calc_mre_K: 0.4144 - val_loss: 70.8987 - val_calc_mre_K: 0.4333\n", "Epoch 1490/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.8794 - calc_mre_K: 0.4143 - val_loss: 69.3453 - val_calc_mre_K: 0.4238\n", "Epoch 1491/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.6992 - calc_mre_K: 0.4132 - val_loss: 71.5107 - val_calc_mre_K: 0.4370\n", "Epoch 1492/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.9070 - calc_mre_K: 0.4145 - val_loss: 68.7203 - val_calc_mre_K: 0.4200\n", "Epoch 1493/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.0223 - calc_mre_K: 0.4152 - val_loss: 69.5239 - val_calc_mre_K: 0.4249\n", "Epoch 1494/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.0333 - calc_mre_K: 0.4152 - val_loss: 71.1869 - val_calc_mre_K: 0.4351\n", "Epoch 1495/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 68.0539 - calc_mre_K: 0.4154 - val_loss: 68.0561 - val_calc_mre_K: 0.4159\n", "Epoch 1496/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 68.0217 - calc_mre_K: 0.4152 - val_loss: 66.6589 - val_calc_mre_K: 0.4074\n", "Epoch 1497/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.8832 - calc_mre_K: 0.4143 - val_loss: 70.3368 - val_calc_mre_K: 0.4299\n", "Epoch 1498/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.0311 - calc_mre_K: 0.4152 - val_loss: 66.7421 - val_calc_mre_K: 0.4079\n", "Epoch 1499/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.7067 - calc_mre_K: 0.4132 - val_loss: 68.4633 - val_calc_mre_K: 0.4184\n", "Epoch 1500/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.7701 - calc_mre_K: 0.4136 - val_loss: 66.8944 - val_calc_mre_K: 0.4088\n", "Epoch 1501/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.0072 - calc_mre_K: 0.4151 - val_loss: 70.3796 - val_calc_mre_K: 0.4301\n", "Epoch 1502/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.3747 - calc_mre_K: 0.4173 - val_loss: 69.8718 - val_calc_mre_K: 0.4270\n", "Epoch 1503/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 68.0534 - calc_mre_K: 0.4154 - val_loss: 65.7478 - val_calc_mre_K: 0.4018\n", "Epoch 1504/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.6125 - calc_mre_K: 0.4127 - val_loss: 67.1617 - val_calc_mre_K: 0.4105\n", "Epoch 1505/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.5160 - calc_mre_K: 0.4121 - val_loss: 66.4056 - val_calc_mre_K: 0.4058\n", "Epoch 1506/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.1968 - calc_mre_K: 0.4162 - val_loss: 66.2801 - val_calc_mre_K: 0.4051\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1507/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.7633 - calc_mre_K: 0.4136 - val_loss: 69.6324 - val_calc_mre_K: 0.4256\n", "Epoch 1508/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.3970 - calc_mre_K: 0.4175 - val_loss: 66.0512 - val_calc_mre_K: 0.4037\n", "Epoch 1509/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.8626 - calc_mre_K: 0.4142 - val_loss: 68.3346 - val_calc_mre_K: 0.4176\n", "Epoch 1510/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.4852 - calc_mre_K: 0.4119 - val_loss: 67.6508 - val_calc_mre_K: 0.4134\n", "Epoch 1511/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.5109 - calc_mre_K: 0.4121 - val_loss: 65.8801 - val_calc_mre_K: 0.4026\n", "Epoch 1512/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.7942 - calc_mre_K: 0.4138 - val_loss: 67.3580 - val_calc_mre_K: 0.4117\n", "Epoch 1513/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.3153 - calc_mre_K: 0.4109 - val_loss: 67.4349 - val_calc_mre_K: 0.4121\n", "Epoch 1514/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.0647 - calc_mre_K: 0.4154 - val_loss: 67.2136 - val_calc_mre_K: 0.4108\n", "Epoch 1515/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 67.5334 - calc_mre_K: 0.4122 - val_loss: 66.2822 - val_calc_mre_K: 0.4051\n", "Epoch 1516/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.8981 - calc_mre_K: 0.4144 - val_loss: 65.8460 - val_calc_mre_K: 0.4024\n", "Epoch 1517/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 68.1698 - calc_mre_K: 0.4161 - val_loss: 67.1923 - val_calc_mre_K: 0.4106\n", "Epoch 1518/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.6026 - calc_mre_K: 0.4126 - val_loss: 72.4937 - val_calc_mre_K: 0.4431\n", "Epoch 1519/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.7897 - calc_mre_K: 0.4138 - val_loss: 67.1793 - val_calc_mre_K: 0.4106\n", "Epoch 1520/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.9709 - calc_mre_K: 0.4149 - val_loss: 74.4675 - val_calc_mre_K: 0.4551\n", "Epoch 1521/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.7952 - calc_mre_K: 0.4138 - val_loss: 67.8982 - val_calc_mre_K: 0.4150\n", "Epoch 1522/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.6228 - calc_mre_K: 0.4127 - val_loss: 68.8949 - val_calc_mre_K: 0.4211\n", "Epoch 1523/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.7469 - calc_mre_K: 0.4135 - val_loss: 66.7173 - val_calc_mre_K: 0.4078\n", "Epoch 1524/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.7330 - calc_mre_K: 0.4134 - val_loss: 71.6530 - val_calc_mre_K: 0.4379\n", "Epoch 1525/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 68.0212 - calc_mre_K: 0.4152 - val_loss: 68.8909 - val_calc_mre_K: 0.4210\n", "Epoch 1526/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.3322 - calc_mre_K: 0.4110 - val_loss: 66.5561 - val_calc_mre_K: 0.4068\n", "Epoch 1527/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 67.3006 - calc_mre_K: 0.4108 - val_loss: 64.4538 - val_calc_mre_K: 0.3939\n", "Epoch 1528/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.8533 - calc_mre_K: 0.4141 - val_loss: 70.3852 - val_calc_mre_K: 0.4302\n", "Epoch 1529/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 67.9034 - calc_mre_K: 0.4144 - val_loss: 68.7256 - val_calc_mre_K: 0.4200\n", "Epoch 1530/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.3115 - calc_mre_K: 0.4108 - val_loss: 67.8300 - val_calc_mre_K: 0.4145\n", "Epoch 1531/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.7857 - calc_mre_K: 0.4137 - val_loss: 65.5512 - val_calc_mre_K: 0.4006\n", "Epoch 1532/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.5139 - calc_mre_K: 0.4121 - val_loss: 67.5175 - val_calc_mre_K: 0.4126\n", "Epoch 1533/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.5637 - calc_mre_K: 0.4124 - val_loss: 68.9405 - val_calc_mre_K: 0.4213\n", "Epoch 1534/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.0539 - calc_mre_K: 0.4154 - val_loss: 65.3777 - val_calc_mre_K: 0.3995\n", "Epoch 1535/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.1695 - calc_mre_K: 0.4100 - val_loss: 66.5851 - val_calc_mre_K: 0.4069\n", "Epoch 1536/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.6419 - calc_mre_K: 0.4129 - val_loss: 65.8892 - val_calc_mre_K: 0.4027\n", "Epoch 1537/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.7625 - calc_mre_K: 0.4136 - val_loss: 66.8922 - val_calc_mre_K: 0.4088\n", "Epoch 1538/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 67.8458 - calc_mre_K: 0.4141 - val_loss: 70.6170 - val_calc_mre_K: 0.4316\n", "Epoch 1539/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.2994 - calc_mre_K: 0.4108 - val_loss: 70.8414 - val_calc_mre_K: 0.4330\n", "Epoch 1540/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.4940 - calc_mre_K: 0.4120 - val_loss: 69.8084 - val_calc_mre_K: 0.4266\n", "Epoch 1541/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.6450 - calc_mre_K: 0.4129 - val_loss: 66.7908 - val_calc_mre_K: 0.4082\n", "Epoch 1542/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.7676 - calc_mre_K: 0.4136 - val_loss: 64.5489 - val_calc_mre_K: 0.3945\n", "Epoch 1543/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 68.1260 - calc_mre_K: 0.4158 - val_loss: 67.8807 - val_calc_mre_K: 0.4149\n", "Epoch 1544/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.7047 - calc_mre_K: 0.4132 - val_loss: 65.8640 - val_calc_mre_K: 0.4025\n", "Epoch 1545/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.6475 - calc_mre_K: 0.4068 - val_loss: 67.6671 - val_calc_mre_K: 0.4136\n", "Epoch 1546/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.0450 - calc_mre_K: 0.4153 - val_loss: 69.7912 - val_calc_mre_K: 0.4265\n", "Epoch 1547/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.7216 - calc_mre_K: 0.4133 - val_loss: 65.2731 - val_calc_mre_K: 0.3989\n", "Epoch 1548/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.8706 - calc_mre_K: 0.4081 - val_loss: 72.3021 - val_calc_mre_K: 0.4419\n", "Epoch 1549/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.5687 - calc_mre_K: 0.4124 - val_loss: 67.2489 - val_calc_mre_K: 0.4110\n", "Epoch 1550/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.2449 - calc_mre_K: 0.4104 - val_loss: 65.9245 - val_calc_mre_K: 0.4029\n", "Epoch 1551/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.3665 - calc_mre_K: 0.4112 - val_loss: 79.1155 - val_calc_mre_K: 0.4835\n", "Epoch 1552/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 67.8234 - calc_mre_K: 0.4140 - val_loss: 74.3852 - val_calc_mre_K: 0.4546\n", "Epoch 1553/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.4545 - calc_mre_K: 0.4117 - val_loss: 70.1657 - val_calc_mre_K: 0.4288\n", "Epoch 1554/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.3491 - calc_mre_K: 0.4111 - val_loss: 66.5029 - val_calc_mre_K: 0.4064\n", "Epoch 1555/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 67.3080 - calc_mre_K: 0.4108 - val_loss: 66.3134 - val_calc_mre_K: 0.4053\n", "Epoch 1556/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 68.2204 - calc_mre_K: 0.4164 - val_loss: 68.5300 - val_calc_mre_K: 0.4188\n", "Epoch 1557/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.1189 - calc_mre_K: 0.4097 - val_loss: 66.9962 - val_calc_mre_K: 0.4094\n", "Epoch 1558/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.6026 - calc_mre_K: 0.4126 - val_loss: 66.4459 - val_calc_mre_K: 0.4061\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1559/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.9909 - calc_mre_K: 0.4089 - val_loss: 71.3867 - val_calc_mre_K: 0.4363\n", "Epoch 1560/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 67.1787 - calc_mre_K: 0.4100 - val_loss: 67.4311 - val_calc_mre_K: 0.4121\n", "Epoch 1561/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 67.1514 - calc_mre_K: 0.4099 - val_loss: 67.0491 - val_calc_mre_K: 0.4098\n", "Epoch 1562/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.7027 - calc_mre_K: 0.4132 - val_loss: 68.0611 - val_calc_mre_K: 0.4160\n", "Epoch 1563/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.2323 - calc_mre_K: 0.4104 - val_loss: 65.2364 - val_calc_mre_K: 0.3987\n", "Epoch 1564/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.0675 - calc_mre_K: 0.4093 - val_loss: 66.6293 - val_calc_mre_K: 0.4072\n", "Epoch 1565/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 68.0607 - calc_mre_K: 0.4154 - val_loss: 66.5293 - val_calc_mre_K: 0.4066\n", "Epoch 1566/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 66.8073 - calc_mre_K: 0.4078 - val_loss: 65.7095 - val_calc_mre_K: 0.4016\n", "Epoch 1567/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.2443 - calc_mre_K: 0.4104 - val_loss: 68.5357 - val_calc_mre_K: 0.4189\n", "Epoch 1568/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.2242 - calc_mre_K: 0.4103 - val_loss: 67.0396 - val_calc_mre_K: 0.4097\n", "Epoch 1569/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.9691 - calc_mre_K: 0.4087 - val_loss: 67.0134 - val_calc_mre_K: 0.4096\n", "Epoch 1570/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.7493 - calc_mre_K: 0.4135 - val_loss: 71.6984 - val_calc_mre_K: 0.4382\n", "Epoch 1571/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.3758 - calc_mre_K: 0.4112 - val_loss: 70.6814 - val_calc_mre_K: 0.4320\n", "Epoch 1572/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.4059 - calc_mre_K: 0.4114 - val_loss: 70.8017 - val_calc_mre_K: 0.4327\n", "Epoch 1573/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 67.0912 - calc_mre_K: 0.4095 - val_loss: 67.1896 - val_calc_mre_K: 0.4106\n", "Epoch 1574/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.0990 - calc_mre_K: 0.4095 - val_loss: 65.7239 - val_calc_mre_K: 0.4017\n", "Epoch 1575/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.9299 - calc_mre_K: 0.4085 - val_loss: 67.1318 - val_calc_mre_K: 0.4103\n", "Epoch 1576/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.5996 - calc_mre_K: 0.4126 - val_loss: 67.2229 - val_calc_mre_K: 0.4108\n", "Epoch 1577/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.7644 - calc_mre_K: 0.4136 - val_loss: 66.3171 - val_calc_mre_K: 0.4053\n", "Epoch 1578/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.9166 - calc_mre_K: 0.4084 - val_loss: 64.8713 - val_calc_mre_K: 0.3965\n", "Epoch 1579/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 68.0201 - calc_mre_K: 0.4152 - val_loss: 67.7652 - val_calc_mre_K: 0.4142\n", "Epoch 1580/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.1016 - calc_mre_K: 0.4096 - val_loss: 70.8176 - val_calc_mre_K: 0.4328\n", "Epoch 1581/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 67.0813 - calc_mre_K: 0.4094 - val_loss: 68.7005 - val_calc_mre_K: 0.4199\n", "Epoch 1582/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.5597 - calc_mre_K: 0.4124 - val_loss: 66.3359 - val_calc_mre_K: 0.4054\n", "Epoch 1583/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.1805 - calc_mre_K: 0.4100 - val_loss: 66.7109 - val_calc_mre_K: 0.4077\n", "Epoch 1584/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.9846 - calc_mre_K: 0.4088 - val_loss: 68.5721 - val_calc_mre_K: 0.4191\n", "Epoch 1585/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.9255 - calc_mre_K: 0.4085 - val_loss: 69.6557 - val_calc_mre_K: 0.4257\n", "Epoch 1586/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.6813 - calc_mre_K: 0.4131 - val_loss: 67.5992 - val_calc_mre_K: 0.4131\n", "Epoch 1587/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.5295 - calc_mre_K: 0.4061 - val_loss: 66.2507 - val_calc_mre_K: 0.4049\n", "Epoch 1588/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.4734 - calc_mre_K: 0.4118 - val_loss: 68.8908 - val_calc_mre_K: 0.4210\n", "Epoch 1589/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.8750 - calc_mre_K: 0.4082 - val_loss: 66.3439 - val_calc_mre_K: 0.4055\n", "Epoch 1590/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.4480 - calc_mre_K: 0.4117 - val_loss: 67.5702 - val_calc_mre_K: 0.4130\n", "Epoch 1591/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 67.3766 - calc_mre_K: 0.4112 - val_loss: 68.7601 - val_calc_mre_K: 0.4202\n", "Epoch 1592/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.9570 - calc_mre_K: 0.4087 - val_loss: 65.4352 - val_calc_mre_K: 0.3999\n", "Epoch 1593/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.0895 - calc_mre_K: 0.4095 - val_loss: 68.3880 - val_calc_mre_K: 0.4180\n", "Epoch 1594/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.8664 - calc_mre_K: 0.4081 - val_loss: 67.9841 - val_calc_mre_K: 0.4155\n", "Epoch 1595/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.8210 - calc_mre_K: 0.4139 - val_loss: 66.2175 - val_calc_mre_K: 0.4047\n", "Epoch 1596/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.7402 - calc_mre_K: 0.4073 - val_loss: 68.1329 - val_calc_mre_K: 0.4164\n", "Epoch 1597/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.6729 - calc_mre_K: 0.4069 - val_loss: 66.7290 - val_calc_mre_K: 0.4078\n", "Epoch 1598/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 67.2740 - calc_mre_K: 0.4106 - val_loss: 67.7112 - val_calc_mre_K: 0.4138\n", "Epoch 1599/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.5131 - calc_mre_K: 0.4060 - val_loss: 73.7308 - val_calc_mre_K: 0.4506\n", "Epoch 1600/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 67.7228 - calc_mre_K: 0.4133 - val_loss: 68.1893 - val_calc_mre_K: 0.4167\n", "Epoch 1601/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.3413 - calc_mre_K: 0.4049 - val_loss: 66.3762 - val_calc_mre_K: 0.4057\n", "Epoch 1602/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 67.1550 - calc_mre_K: 0.4099 - val_loss: 71.2975 - val_calc_mre_K: 0.4358\n", "Epoch 1603/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.9676 - calc_mre_K: 0.4087 - val_loss: 67.6279 - val_calc_mre_K: 0.4133\n", "Epoch 1604/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.4937 - calc_mre_K: 0.4058 - val_loss: 67.5114 - val_calc_mre_K: 0.4126\n", "Epoch 1605/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.4115 - calc_mre_K: 0.4114 - val_loss: 71.7585 - val_calc_mre_K: 0.4386\n", "Epoch 1606/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.4128 - calc_mre_K: 0.4115 - val_loss: 67.4315 - val_calc_mre_K: 0.4121\n", "Epoch 1607/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 66.7711 - calc_mre_K: 0.4075 - val_loss: 67.2188 - val_calc_mre_K: 0.4108\n", "Epoch 1608/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.2398 - calc_mre_K: 0.4104 - val_loss: 66.9243 - val_calc_mre_K: 0.4090\n", "Epoch 1609/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.4228 - calc_mre_K: 0.4054 - val_loss: 67.5207 - val_calc_mre_K: 0.4126\n", "Epoch 1610/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 67.5993 - calc_mre_K: 0.4126 - val_loss: 71.0532 - val_calc_mre_K: 0.4342\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1611/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.7956 - calc_mre_K: 0.4077 - val_loss: 65.8377 - val_calc_mre_K: 0.4024\n", "Epoch 1612/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.2697 - calc_mre_K: 0.4106 - val_loss: 75.4200 - val_calc_mre_K: 0.4609\n", "Epoch 1613/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 66.6555 - calc_mre_K: 0.4068 - val_loss: 66.2911 - val_calc_mre_K: 0.4051\n", "Epoch 1614/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 67.3950 - calc_mre_K: 0.4113 - val_loss: 64.9577 - val_calc_mre_K: 0.3970\n", "Epoch 1615/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.8341 - calc_mre_K: 0.4079 - val_loss: 64.7493 - val_calc_mre_K: 0.3957\n", "Epoch 1616/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 67.0432 - calc_mre_K: 0.4092 - val_loss: 69.5972 - val_calc_mre_K: 0.4253\n", "Epoch 1617/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.1712 - calc_mre_K: 0.4100 - val_loss: 65.5221 - val_calc_mre_K: 0.4004\n", "Epoch 1618/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.7514 - calc_mre_K: 0.4074 - val_loss: 64.9627 - val_calc_mre_K: 0.3970\n", "Epoch 1619/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.5372 - calc_mre_K: 0.4061 - val_loss: 66.8253 - val_calc_mre_K: 0.4084\n", "Epoch 1620/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.9560 - calc_mre_K: 0.4087 - val_loss: 68.6696 - val_calc_mre_K: 0.4197\n", "Epoch 1621/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 67.4241 - calc_mre_K: 0.4115 - val_loss: 64.6679 - val_calc_mre_K: 0.3952\n", "Epoch 1622/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.0546 - calc_mre_K: 0.4093 - val_loss: 65.5165 - val_calc_mre_K: 0.4004\n", "Epoch 1623/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.8029 - calc_mre_K: 0.4077 - val_loss: 69.1462 - val_calc_mre_K: 0.4226\n", "Epoch 1624/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.6231 - calc_mre_K: 0.4066 - val_loss: 68.3552 - val_calc_mre_K: 0.4178\n", "Epoch 1625/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.4747 - calc_mre_K: 0.4057 - val_loss: 67.1411 - val_calc_mre_K: 0.4104\n", "Epoch 1626/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.0215 - calc_mre_K: 0.4091 - val_loss: 68.8543 - val_calc_mre_K: 0.4208\n", "Epoch 1627/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.0263 - calc_mre_K: 0.4091 - val_loss: 69.5473 - val_calc_mre_K: 0.4250\n", "Epoch 1628/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.1334 - calc_mre_K: 0.4097 - val_loss: 65.8345 - val_calc_mre_K: 0.4024\n", "Epoch 1629/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.7709 - calc_mre_K: 0.4075 - val_loss: 65.2146 - val_calc_mre_K: 0.3986\n", "Epoch 1630/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 67.1541 - calc_mre_K: 0.4099 - val_loss: 66.5847 - val_calc_mre_K: 0.4069\n", "Epoch 1631/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.4499 - calc_mre_K: 0.4056 - val_loss: 67.1843 - val_calc_mre_K: 0.4106\n", "Epoch 1632/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.9526 - calc_mre_K: 0.4086 - val_loss: 65.9973 - val_calc_mre_K: 0.4034\n", "Epoch 1633/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.1600 - calc_mre_K: 0.4099 - val_loss: 67.0773 - val_calc_mre_K: 0.4100\n", "Epoch 1634/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.6078 - calc_mre_K: 0.4065 - val_loss: 65.8972 - val_calc_mre_K: 0.4027\n", "Epoch 1635/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.5592 - calc_mre_K: 0.4062 - val_loss: 66.4557 - val_calc_mre_K: 0.4062\n", "Epoch 1636/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 67.0091 - calc_mre_K: 0.4090 - val_loss: 69.2700 - val_calc_mre_K: 0.4233\n", "Epoch 1637/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.0707 - calc_mre_K: 0.4033 - val_loss: 67.7917 - val_calc_mre_K: 0.4143\n", "Epoch 1638/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.8072 - calc_mre_K: 0.4078 - val_loss: 65.5513 - val_calc_mre_K: 0.4006\n", "Epoch 1639/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.9468 - calc_mre_K: 0.4086 - val_loss: 67.9053 - val_calc_mre_K: 0.4150\n", "Epoch 1640/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 66.3346 - calc_mre_K: 0.4049 - val_loss: 68.9369 - val_calc_mre_K: 0.4213\n", "Epoch 1641/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.7940 - calc_mre_K: 0.4077 - val_loss: 65.7871 - val_calc_mre_K: 0.4021\n", "Epoch 1642/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.1277 - calc_mre_K: 0.4097 - val_loss: 67.6338 - val_calc_mre_K: 0.4134\n", "Epoch 1643/2000\n", "48000/48000 [==============================] - 3s 62us/step - loss: 66.6223 - calc_mre_K: 0.4066 - val_loss: 67.6507 - val_calc_mre_K: 0.4134\n", "Epoch 1644/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.0015 - calc_mre_K: 0.4089 - val_loss: 66.8031 - val_calc_mre_K: 0.4083\n", "Epoch 1645/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 65.9968 - calc_mre_K: 0.4028 - val_loss: 64.6271 - val_calc_mre_K: 0.3950\n", "Epoch 1646/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.3911 - calc_mre_K: 0.4052 - val_loss: 72.1022 - val_calc_mre_K: 0.4407\n", "Epoch 1647/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 67.3215 - calc_mre_K: 0.4109 - val_loss: 66.2862 - val_calc_mre_K: 0.4051\n", "Epoch 1648/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.8669 - calc_mre_K: 0.4081 - val_loss: 67.1568 - val_calc_mre_K: 0.4104\n", "Epoch 1649/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 66.7773 - calc_mre_K: 0.4076 - val_loss: 64.7445 - val_calc_mre_K: 0.3957\n", "Epoch 1650/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.5722 - calc_mre_K: 0.4063 - val_loss: 75.7173 - val_calc_mre_K: 0.4628\n", "Epoch 1651/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.5493 - calc_mre_K: 0.4062 - val_loss: 65.6982 - val_calc_mre_K: 0.4015\n", "Epoch 1652/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.8130 - calc_mre_K: 0.4078 - val_loss: 71.2929 - val_calc_mre_K: 0.4357\n", "Epoch 1653/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.5455 - calc_mre_K: 0.4062 - val_loss: 66.7396 - val_calc_mre_K: 0.4079\n", "Epoch 1654/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.5103 - calc_mre_K: 0.4059 - val_loss: 64.2608 - val_calc_mre_K: 0.3927\n", "Epoch 1655/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.7599 - calc_mre_K: 0.4075 - val_loss: 66.1260 - val_calc_mre_K: 0.4041\n", "Epoch 1656/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.2375 - calc_mre_K: 0.4043 - val_loss: 66.5685 - val_calc_mre_K: 0.4068\n", "Epoch 1657/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.8528 - calc_mre_K: 0.4080 - val_loss: 69.8497 - val_calc_mre_K: 0.4269\n", "Epoch 1658/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.6901 - calc_mre_K: 0.4070 - val_loss: 65.2509 - val_calc_mre_K: 0.3988\n", "Epoch 1659/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 67.3340 - calc_mre_K: 0.4110 - val_loss: 67.4362 - val_calc_mre_K: 0.4122\n", "Epoch 1660/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.4083 - calc_mre_K: 0.4053 - val_loss: 68.3998 - val_calc_mre_K: 0.4180\n", "Epoch 1661/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.1948 - calc_mre_K: 0.4040 - val_loss: 68.8213 - val_calc_mre_K: 0.4206\n", "Epoch 1662/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.5560 - calc_mre_K: 0.4062 - val_loss: 68.2103 - val_calc_mre_K: 0.4169\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1663/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 67.1910 - calc_mre_K: 0.4101 - val_loss: 68.0159 - val_calc_mre_K: 0.4157\n", "Epoch 1664/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.0932 - calc_mre_K: 0.4034 - val_loss: 65.4491 - val_calc_mre_K: 0.4000\n", "Epoch 1665/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 67.0873 - calc_mre_K: 0.4095 - val_loss: 66.4690 - val_calc_mre_K: 0.4062\n", "Epoch 1666/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.3738 - calc_mre_K: 0.4051 - val_loss: 67.0305 - val_calc_mre_K: 0.4097\n", "Epoch 1667/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.7392 - calc_mre_K: 0.4073 - val_loss: 67.9774 - val_calc_mre_K: 0.4154\n", "Epoch 1668/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.6158 - calc_mre_K: 0.4066 - val_loss: 65.7546 - val_calc_mre_K: 0.4019\n", "Epoch 1669/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.0146 - calc_mre_K: 0.4029 - val_loss: 64.8097 - val_calc_mre_K: 0.3961\n", "Epoch 1670/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.1517 - calc_mre_K: 0.4099 - val_loss: 63.4925 - val_calc_mre_K: 0.3880\n", "Epoch 1671/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.1469 - calc_mre_K: 0.4037 - val_loss: 64.8505 - val_calc_mre_K: 0.3963\n", "Epoch 1672/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.6094 - calc_mre_K: 0.4066 - val_loss: 67.2998 - val_calc_mre_K: 0.4113\n", "Epoch 1673/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.4631 - calc_mre_K: 0.4057 - val_loss: 68.2488 - val_calc_mre_K: 0.4171\n", "Epoch 1674/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.6624 - calc_mre_K: 0.4069 - val_loss: 66.0231 - val_calc_mre_K: 0.4035\n", "Epoch 1675/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.3323 - calc_mre_K: 0.4049 - val_loss: 64.2077 - val_calc_mre_K: 0.3924\n", "Epoch 1676/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.4942 - calc_mre_K: 0.4058 - val_loss: 65.0185 - val_calc_mre_K: 0.3974\n", "Epoch 1677/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.7086 - calc_mre_K: 0.4072 - val_loss: 65.0017 - val_calc_mre_K: 0.3973\n", "Epoch 1678/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.5455 - calc_mre_K: 0.4062 - val_loss: 65.3553 - val_calc_mre_K: 0.3994\n", "Epoch 1679/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.2581 - calc_mre_K: 0.4044 - val_loss: 72.7300 - val_calc_mre_K: 0.4445\n", "Epoch 1680/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.9411 - calc_mre_K: 0.4025 - val_loss: 63.6383 - val_calc_mre_K: 0.3889\n", "Epoch 1681/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.3580 - calc_mre_K: 0.4050 - val_loss: 71.0317 - val_calc_mre_K: 0.4341\n", "Epoch 1682/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.8865 - calc_mre_K: 0.4082 - val_loss: 63.6460 - val_calc_mre_K: 0.3890\n", "Epoch 1683/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.0875 - calc_mre_K: 0.4034 - val_loss: 69.2292 - val_calc_mre_K: 0.4231\n", "Epoch 1684/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.2744 - calc_mre_K: 0.4045 - val_loss: 64.7165 - val_calc_mre_K: 0.3955\n", "Epoch 1685/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 66.5026 - calc_mre_K: 0.4059 - val_loss: 69.4952 - val_calc_mre_K: 0.4247\n", "Epoch 1686/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 66.1864 - calc_mre_K: 0.4040 - val_loss: 69.7241 - val_calc_mre_K: 0.4261\n", "Epoch 1687/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.3482 - calc_mre_K: 0.4050 - val_loss: 64.0355 - val_calc_mre_K: 0.3914\n", "Epoch 1688/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.3490 - calc_mre_K: 0.4050 - val_loss: 66.5154 - val_calc_mre_K: 0.4065\n", "Epoch 1689/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.5119 - calc_mre_K: 0.4060 - val_loss: 65.8452 - val_calc_mre_K: 0.4024\n", "Epoch 1690/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.6447 - calc_mre_K: 0.4068 - val_loss: 74.2936 - val_calc_mre_K: 0.4541\n", "Epoch 1691/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.6177 - calc_mre_K: 0.4005 - val_loss: 67.9897 - val_calc_mre_K: 0.4155\n", "Epoch 1692/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 67.0255 - calc_mre_K: 0.4091 - val_loss: 67.1729 - val_calc_mre_K: 0.4105\n", "Epoch 1693/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.5474 - calc_mre_K: 0.4062 - val_loss: 83.8997 - val_calc_mre_K: 0.5128\n", "Epoch 1694/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.1134 - calc_mre_K: 0.4035 - val_loss: 69.8472 - val_calc_mre_K: 0.4269\n", "Epoch 1695/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.6123 - calc_mre_K: 0.4066 - val_loss: 65.4202 - val_calc_mre_K: 0.3998\n", "Epoch 1696/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.1564 - calc_mre_K: 0.4038 - val_loss: 68.2759 - val_calc_mre_K: 0.4173\n", "Epoch 1697/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 65.6833 - calc_mre_K: 0.4009 - val_loss: 65.4262 - val_calc_mre_K: 0.3999\n", "Epoch 1698/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.0551 - calc_mre_K: 0.4032 - val_loss: 70.3624 - val_calc_mre_K: 0.4300\n", "Epoch 1699/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.0954 - calc_mre_K: 0.4034 - val_loss: 64.5018 - val_calc_mre_K: 0.3942\n", "Epoch 1700/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 65.8725 - calc_mre_K: 0.4021 - val_loss: 65.2802 - val_calc_mre_K: 0.3990\n", "Epoch 1701/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.3755 - calc_mre_K: 0.4051 - val_loss: 65.2134 - val_calc_mre_K: 0.3986\n", "Epoch 1702/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.3429 - calc_mre_K: 0.4049 - val_loss: 67.2513 - val_calc_mre_K: 0.4110\n", "Epoch 1703/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.4103 - calc_mre_K: 0.4053 - val_loss: 64.2614 - val_calc_mre_K: 0.3927\n", "Epoch 1704/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.1269 - calc_mre_K: 0.4036 - val_loss: 67.2453 - val_calc_mre_K: 0.4110\n", "Epoch 1705/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.7234 - calc_mre_K: 0.4011 - val_loss: 67.2578 - val_calc_mre_K: 0.4111\n", "Epoch 1706/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.2383 - calc_mre_K: 0.4043 - val_loss: 63.8773 - val_calc_mre_K: 0.3904\n", "Epoch 1707/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.0282 - calc_mre_K: 0.4030 - val_loss: 69.1036 - val_calc_mre_K: 0.4223\n", "Epoch 1708/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.6667 - calc_mre_K: 0.4069 - val_loss: 68.2404 - val_calc_mre_K: 0.4171\n", "Epoch 1709/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 65.8336 - calc_mre_K: 0.4018 - val_loss: 64.9424 - val_calc_mre_K: 0.3969\n", "Epoch 1710/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.8459 - calc_mre_K: 0.4019 - val_loss: 65.1709 - val_calc_mre_K: 0.3983\n", "Epoch 1711/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.2193 - calc_mre_K: 0.4042 - val_loss: 67.8393 - val_calc_mre_K: 0.4146\n", "Epoch 1712/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.7206 - calc_mre_K: 0.4011 - val_loss: 68.8724 - val_calc_mre_K: 0.4209\n", "Epoch 1713/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.7116 - calc_mre_K: 0.4072 - val_loss: 70.2543 - val_calc_mre_K: 0.4294\n", "Epoch 1714/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.9383 - calc_mre_K: 0.4025 - val_loss: 64.9953 - val_calc_mre_K: 0.3972\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1715/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.7120 - calc_mre_K: 0.4011 - val_loss: 69.3187 - val_calc_mre_K: 0.4236\n", "Epoch 1716/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.5953 - calc_mre_K: 0.4004 - val_loss: 65.6914 - val_calc_mre_K: 0.4015\n", "Epoch 1717/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.3623 - calc_mre_K: 0.4050 - val_loss: 65.9692 - val_calc_mre_K: 0.4032\n", "Epoch 1718/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.0976 - calc_mre_K: 0.4034 - val_loss: 68.4511 - val_calc_mre_K: 0.4184\n", "Epoch 1719/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.5417 - calc_mre_K: 0.4061 - val_loss: 64.5129 - val_calc_mre_K: 0.3943\n", "Epoch 1720/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 66.1490 - calc_mre_K: 0.4037 - val_loss: 67.7093 - val_calc_mre_K: 0.4138\n", "Epoch 1721/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.0594 - calc_mre_K: 0.4032 - val_loss: 62.8483 - val_calc_mre_K: 0.3841\n", "Epoch 1722/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.7840 - calc_mre_K: 0.4015 - val_loss: 69.4149 - val_calc_mre_K: 0.4242\n", "Epoch 1723/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 65.9073 - calc_mre_K: 0.4023 - val_loss: 67.0201 - val_calc_mre_K: 0.4096\n", "Epoch 1724/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.2140 - calc_mre_K: 0.4041 - val_loss: 66.8258 - val_calc_mre_K: 0.4084\n", "Epoch 1725/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.1535 - calc_mre_K: 0.4038 - val_loss: 68.8263 - val_calc_mre_K: 0.4206\n", "Epoch 1726/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.0479 - calc_mre_K: 0.4031 - val_loss: 68.5411 - val_calc_mre_K: 0.4189\n", "Epoch 1727/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.7465 - calc_mre_K: 0.4013 - val_loss: 64.6073 - val_calc_mre_K: 0.3949\n", "Epoch 1728/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 66.0574 - calc_mre_K: 0.4032 - val_loss: 64.3645 - val_calc_mre_K: 0.3934\n", "Epoch 1729/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 65.9239 - calc_mre_K: 0.4024 - val_loss: 66.8568 - val_calc_mre_K: 0.4086\n", "Epoch 1730/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.1229 - calc_mre_K: 0.4036 - val_loss: 64.8373 - val_calc_mre_K: 0.3963\n", "Epoch 1731/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.9799 - calc_mre_K: 0.4027 - val_loss: 65.5623 - val_calc_mre_K: 0.4007\n", "Epoch 1732/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.3137 - calc_mre_K: 0.4047 - val_loss: 62.6383 - val_calc_mre_K: 0.3828\n", "Epoch 1733/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 65.5847 - calc_mre_K: 0.4003 - val_loss: 67.3954 - val_calc_mre_K: 0.4119\n", "Epoch 1734/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 66.0212 - calc_mre_K: 0.4030 - val_loss: 68.1371 - val_calc_mre_K: 0.4164\n", "Epoch 1735/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 65.6356 - calc_mre_K: 0.4006 - val_loss: 63.7656 - val_calc_mre_K: 0.3897\n", "Epoch 1736/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 65.9332 - calc_mre_K: 0.4024 - val_loss: 66.5565 - val_calc_mre_K: 0.4068\n", "Epoch 1737/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 65.4161 - calc_mre_K: 0.3993 - val_loss: 65.8031 - val_calc_mre_K: 0.4022\n", "Epoch 1738/2000\n", "48000/48000 [==============================] - 3s 61us/step - loss: 65.6400 - calc_mre_K: 0.4006 - val_loss: 64.3310 - val_calc_mre_K: 0.3932\n", "Epoch 1739/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.9933 - calc_mre_K: 0.4028 - val_loss: 66.4963 - val_calc_mre_K: 0.4064\n", "Epoch 1740/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 66.1581 - calc_mre_K: 0.4038 - val_loss: 63.6453 - val_calc_mre_K: 0.3890\n", "Epoch 1741/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 66.1498 - calc_mre_K: 0.4037 - val_loss: 72.3979 - val_calc_mre_K: 0.4425\n", "Epoch 1742/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 65.6090 - calc_mre_K: 0.4004 - val_loss: 66.1453 - val_calc_mre_K: 0.4043\n", "Epoch 1743/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 65.8689 - calc_mre_K: 0.4020 - val_loss: 67.7365 - val_calc_mre_K: 0.4140\n", "Epoch 1744/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.5131 - calc_mre_K: 0.4060 - val_loss: 64.7236 - val_calc_mre_K: 0.3956\n", "Epoch 1745/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.5918 - calc_mre_K: 0.4003 - val_loss: 67.5573 - val_calc_mre_K: 0.4129\n", "Epoch 1746/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.8291 - calc_mre_K: 0.4018 - val_loss: 65.6058 - val_calc_mre_K: 0.4010\n", "Epoch 1747/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.2168 - calc_mre_K: 0.3981 - val_loss: 64.2847 - val_calc_mre_K: 0.3929\n", "Epoch 1748/2000\n", "48000/48000 [==============================] - 3s 70us/step - loss: 65.2212 - calc_mre_K: 0.3981 - val_loss: 65.1446 - val_calc_mre_K: 0.3981\n", "Epoch 1749/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 66.0066 - calc_mre_K: 0.4029 - val_loss: 74.8873 - val_calc_mre_K: 0.4577\n", "Epoch 1750/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.6105 - calc_mre_K: 0.4005 - val_loss: 66.9214 - val_calc_mre_K: 0.4090\n", "Epoch 1751/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.7220 - calc_mre_K: 0.4011 - val_loss: 68.8323 - val_calc_mre_K: 0.4207\n", "Epoch 1752/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.4400 - calc_mre_K: 0.3994 - val_loss: 66.0599 - val_calc_mre_K: 0.4037\n", "Epoch 1753/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 65.3779 - calc_mre_K: 0.3990 - val_loss: 71.2653 - val_calc_mre_K: 0.4355\n", "Epoch 1754/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 66.6758 - calc_mre_K: 0.4070 - val_loss: 74.7734 - val_calc_mre_K: 0.4570\n", "Epoch 1755/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.0944 - calc_mre_K: 0.3973 - val_loss: 64.8423 - val_calc_mre_K: 0.3963\n", "Epoch 1756/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.7024 - calc_mre_K: 0.4010 - val_loss: 65.6006 - val_calc_mre_K: 0.4009\n", "Epoch 1757/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.2103 - calc_mre_K: 0.3980 - val_loss: 67.0639 - val_calc_mre_K: 0.4099\n", "Epoch 1758/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.5894 - calc_mre_K: 0.4003 - val_loss: 66.2453 - val_calc_mre_K: 0.4049\n", "Epoch 1759/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.8645 - calc_mre_K: 0.4020 - val_loss: 68.7731 - val_calc_mre_K: 0.4203\n", "Epoch 1760/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.2821 - calc_mre_K: 0.3985 - val_loss: 63.9303 - val_calc_mre_K: 0.3907\n", "Epoch 1761/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 66.1436 - calc_mre_K: 0.4037 - val_loss: 63.9247 - val_calc_mre_K: 0.3907\n", "Epoch 1762/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 66.4288 - calc_mre_K: 0.4054 - val_loss: 68.2090 - val_calc_mre_K: 0.4169\n", "Epoch 1763/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.9053 - calc_mre_K: 0.3962 - val_loss: 64.0541 - val_calc_mre_K: 0.3915\n", "Epoch 1764/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.0752 - calc_mre_K: 0.3972 - val_loss: 66.2393 - val_calc_mre_K: 0.4048\n", "Epoch 1765/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 66.0382 - calc_mre_K: 0.4031 - val_loss: 70.7126 - val_calc_mre_K: 0.4322\n", "Epoch 1766/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.2416 - calc_mre_K: 0.3982 - val_loss: 66.4057 - val_calc_mre_K: 0.4058\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1767/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.4979 - calc_mre_K: 0.3998 - val_loss: 63.3799 - val_calc_mre_K: 0.3874\n", "Epoch 1768/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 65.0213 - calc_mre_K: 0.3969 - val_loss: 64.5083 - val_calc_mre_K: 0.3942\n", "Epoch 1769/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.2521 - calc_mre_K: 0.3983 - val_loss: 65.6412 - val_calc_mre_K: 0.4012\n", "Epoch 1770/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.7377 - calc_mre_K: 0.4012 - val_loss: 65.1243 - val_calc_mre_K: 0.3980\n", "Epoch 1771/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.3049 - calc_mre_K: 0.3986 - val_loss: 69.4051 - val_calc_mre_K: 0.4242\n", "Epoch 1772/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.1342 - calc_mre_K: 0.3975 - val_loss: 64.5752 - val_calc_mre_K: 0.3947\n", "Epoch 1773/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.6073 - calc_mre_K: 0.4004 - val_loss: 70.4014 - val_calc_mre_K: 0.4303\n", "Epoch 1774/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.3832 - calc_mre_K: 0.3991 - val_loss: 68.8498 - val_calc_mre_K: 0.4208\n", "Epoch 1775/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.5499 - calc_mre_K: 0.4001 - val_loss: 62.2524 - val_calc_mre_K: 0.3805\n", "Epoch 1776/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.8634 - calc_mre_K: 0.3959 - val_loss: 64.9522 - val_calc_mre_K: 0.3970\n", "Epoch 1777/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.4493 - calc_mre_K: 0.3995 - val_loss: 65.8963 - val_calc_mre_K: 0.4027\n", "Epoch 1778/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.2379 - calc_mre_K: 0.3982 - val_loss: 65.8758 - val_calc_mre_K: 0.4026\n", "Epoch 1779/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.6215 - calc_mre_K: 0.4005 - val_loss: 66.0607 - val_calc_mre_K: 0.4037\n", "Epoch 1780/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.0407 - calc_mre_K: 0.3970 - val_loss: 67.7498 - val_calc_mre_K: 0.4141\n", "Epoch 1781/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.2934 - calc_mre_K: 0.3985 - val_loss: 64.0222 - val_calc_mre_K: 0.3913\n", "Epoch 1782/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.5711 - calc_mre_K: 0.4002 - val_loss: 76.7151 - val_calc_mre_K: 0.4689\n", "Epoch 1783/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.0981 - calc_mre_K: 0.3973 - val_loss: 67.2591 - val_calc_mre_K: 0.4111\n", "Epoch 1784/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.1038 - calc_mre_K: 0.3974 - val_loss: 66.9541 - val_calc_mre_K: 0.4092\n", "Epoch 1785/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 65.2290 - calc_mre_K: 0.3981 - val_loss: 64.8771 - val_calc_mre_K: 0.3965\n", "Epoch 1786/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.0001 - calc_mre_K: 0.3967 - val_loss: 64.1358 - val_calc_mre_K: 0.3920\n", "Epoch 1787/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.3208 - calc_mre_K: 0.3987 - val_loss: 63.8674 - val_calc_mre_K: 0.3903\n", "Epoch 1788/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.5537 - calc_mre_K: 0.4001 - val_loss: 63.9030 - val_calc_mre_K: 0.3905\n", "Epoch 1789/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 65.0985 - calc_mre_K: 0.3973 - val_loss: 67.6849 - val_calc_mre_K: 0.4137\n", "Epoch 1790/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.1348 - calc_mre_K: 0.3976 - val_loss: 63.6331 - val_calc_mre_K: 0.3889\n", "Epoch 1791/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.2417 - calc_mre_K: 0.3982 - val_loss: 67.9837 - val_calc_mre_K: 0.4155\n", "Epoch 1792/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.8970 - calc_mre_K: 0.3961 - val_loss: 62.8854 - val_calc_mre_K: 0.3843\n", "Epoch 1793/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 65.8983 - calc_mre_K: 0.4022 - val_loss: 63.2403 - val_calc_mre_K: 0.3865\n", "Epoch 1794/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 65.0895 - calc_mre_K: 0.3973 - val_loss: 63.0476 - val_calc_mre_K: 0.3853\n", "Epoch 1795/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.7603 - calc_mre_K: 0.3953 - val_loss: 68.0679 - val_calc_mre_K: 0.4160\n", "Epoch 1796/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.1009 - calc_mre_K: 0.3973 - val_loss: 63.4714 - val_calc_mre_K: 0.3879\n", "Epoch 1797/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.7972 - calc_mre_K: 0.3955 - val_loss: 65.9541 - val_calc_mre_K: 0.4031\n", "Epoch 1798/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.1304 - calc_mre_K: 0.3975 - val_loss: 64.6206 - val_calc_mre_K: 0.3949\n", "Epoch 1799/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 65.1303 - calc_mre_K: 0.3975 - val_loss: 69.1191 - val_calc_mre_K: 0.4224\n", "Epoch 1800/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 64.8681 - calc_mre_K: 0.3959 - val_loss: 66.4154 - val_calc_mre_K: 0.4059\n", "Epoch 1801/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 64.7524 - calc_mre_K: 0.3952 - val_loss: 63.6903 - val_calc_mre_K: 0.3892\n", "Epoch 1802/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.1353 - calc_mre_K: 0.3976 - val_loss: 62.9939 - val_calc_mre_K: 0.3850\n", "Epoch 1803/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.6773 - calc_mre_K: 0.4009 - val_loss: 64.0070 - val_calc_mre_K: 0.3912\n", "Epoch 1804/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.4940 - calc_mre_K: 0.3936 - val_loss: 62.7505 - val_calc_mre_K: 0.3835\n", "Epoch 1805/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.1656 - calc_mre_K: 0.3916 - val_loss: 67.5587 - val_calc_mre_K: 0.4129\n", "Epoch 1806/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 65.7763 - calc_mre_K: 0.4015 - val_loss: 67.7273 - val_calc_mre_K: 0.4139\n", "Epoch 1807/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.5917 - calc_mre_K: 0.4003 - val_loss: 63.2748 - val_calc_mre_K: 0.3867\n", "Epoch 1808/2000\n", "48000/48000 [==============================] - 3s 68us/step - loss: 64.6871 - calc_mre_K: 0.3948 - val_loss: 64.4588 - val_calc_mre_K: 0.3940\n", "Epoch 1809/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 65.0048 - calc_mre_K: 0.3968 - val_loss: 64.7940 - val_calc_mre_K: 0.3960\n", "Epoch 1810/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.3931 - calc_mre_K: 0.3991 - val_loss: 63.0132 - val_calc_mre_K: 0.3851\n", "Epoch 1811/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.8673 - calc_mre_K: 0.3959 - val_loss: 63.9935 - val_calc_mre_K: 0.3911\n", "Epoch 1812/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.2704 - calc_mre_K: 0.3923 - val_loss: 62.7574 - val_calc_mre_K: 0.3836\n", "Epoch 1813/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.3618 - calc_mre_K: 0.3989 - val_loss: 62.2387 - val_calc_mre_K: 0.3804\n", "Epoch 1814/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.7062 - calc_mre_K: 0.3949 - val_loss: 64.2283 - val_calc_mre_K: 0.3925\n", "Epoch 1815/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 64.5799 - calc_mre_K: 0.3942 - val_loss: 63.0632 - val_calc_mre_K: 0.3854\n", "Epoch 1816/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 65.5035 - calc_mre_K: 0.3998 - val_loss: 84.1480 - val_calc_mre_K: 0.5143\n", "Epoch 1817/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.5818 - calc_mre_K: 0.3942 - val_loss: 61.4604 - val_calc_mre_K: 0.3756\n", "Epoch 1818/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 64.9168 - calc_mre_K: 0.3962 - val_loss: 68.2657 - val_calc_mre_K: 0.4172\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1819/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.4830 - calc_mre_K: 0.3936 - val_loss: 63.5701 - val_calc_mre_K: 0.3885\n", "Epoch 1820/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.6469 - calc_mre_K: 0.4007 - val_loss: 65.5887 - val_calc_mre_K: 0.4009\n", "Epoch 1821/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 65.3580 - calc_mre_K: 0.3989 - val_loss: 63.9132 - val_calc_mre_K: 0.3906\n", "Epoch 1822/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.4740 - calc_mre_K: 0.3935 - val_loss: 66.3685 - val_calc_mre_K: 0.4056\n", "Epoch 1823/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 64.9790 - calc_mre_K: 0.3966 - val_loss: 68.0156 - val_calc_mre_K: 0.4157\n", "Epoch 1824/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.8575 - calc_mre_K: 0.3959 - val_loss: 65.9024 - val_calc_mre_K: 0.4028\n", "Epoch 1825/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.5452 - calc_mre_K: 0.3940 - val_loss: 70.7649 - val_calc_mre_K: 0.4325\n", "Epoch 1826/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 65.2387 - calc_mre_K: 0.3982 - val_loss: 64.9778 - val_calc_mre_K: 0.3971\n", "Epoch 1827/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.7359 - calc_mre_K: 0.3951 - val_loss: 62.4778 - val_calc_mre_K: 0.3818\n", "Epoch 1828/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.8562 - calc_mre_K: 0.3959 - val_loss: 70.3937 - val_calc_mre_K: 0.4302\n", "Epoch 1829/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.4942 - calc_mre_K: 0.3997 - val_loss: 64.1770 - val_calc_mre_K: 0.3922\n", "Epoch 1830/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 64.0584 - calc_mre_K: 0.3910 - val_loss: 63.3728 - val_calc_mre_K: 0.3873\n", "Epoch 1831/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.2777 - calc_mre_K: 0.3984 - val_loss: 64.9308 - val_calc_mre_K: 0.3968\n", "Epoch 1832/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.5511 - calc_mre_K: 0.3940 - val_loss: 66.2585 - val_calc_mre_K: 0.4049\n", "Epoch 1833/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.6476 - calc_mre_K: 0.3946 - val_loss: 61.6163 - val_calc_mre_K: 0.3766\n", "Epoch 1834/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.7382 - calc_mre_K: 0.3951 - val_loss: 68.4541 - val_calc_mre_K: 0.4184\n", "Epoch 1835/2000\n", "48000/48000 [==============================] - 2s 49us/step - loss: 65.0346 - calc_mre_K: 0.3969 - val_loss: 62.7429 - val_calc_mre_K: 0.3835\n", "Epoch 1836/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.4394 - calc_mre_K: 0.3933 - val_loss: 65.4213 - val_calc_mre_K: 0.3998\n", "Epoch 1837/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.4628 - calc_mre_K: 0.3934 - val_loss: 62.6210 - val_calc_mre_K: 0.3827\n", "Epoch 1838/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.5690 - calc_mre_K: 0.3941 - val_loss: 67.7839 - val_calc_mre_K: 0.4143\n", "Epoch 1839/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.6994 - calc_mre_K: 0.3949 - val_loss: 62.8830 - val_calc_mre_K: 0.3843\n", "Epoch 1840/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.6342 - calc_mre_K: 0.3945 - val_loss: 62.0102 - val_calc_mre_K: 0.3790\n", "Epoch 1841/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.8767 - calc_mre_K: 0.3960 - val_loss: 64.3081 - val_calc_mre_K: 0.3930\n", "Epoch 1842/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.3741 - calc_mre_K: 0.3929 - val_loss: 64.0844 - val_calc_mre_K: 0.3917\n", "Epoch 1843/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.0332 - calc_mre_K: 0.3969 - val_loss: 65.2594 - val_calc_mre_K: 0.3988\n", "Epoch 1844/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.6374 - calc_mre_K: 0.3945 - val_loss: 64.9827 - val_calc_mre_K: 0.3971\n", "Epoch 1845/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 65.0500 - calc_mre_K: 0.3970 - val_loss: 63.3113 - val_calc_mre_K: 0.3869\n", "Epoch 1846/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.5589 - calc_mre_K: 0.3940 - val_loss: 66.4471 - val_calc_mre_K: 0.4061\n", "Epoch 1847/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.6364 - calc_mre_K: 0.3945 - val_loss: 62.6578 - val_calc_mre_K: 0.3829\n", "Epoch 1848/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.1222 - calc_mre_K: 0.3975 - val_loss: 66.0702 - val_calc_mre_K: 0.4038\n", "Epoch 1849/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.0463 - calc_mre_K: 0.3909 - val_loss: 68.0513 - val_calc_mre_K: 0.4159\n", "Epoch 1850/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.6709 - calc_mre_K: 0.3947 - val_loss: 63.5787 - val_calc_mre_K: 0.3886\n", "Epoch 1851/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.9307 - calc_mre_K: 0.3963 - val_loss: 63.7381 - val_calc_mre_K: 0.3895\n", "Epoch 1852/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.4757 - calc_mre_K: 0.3935 - val_loss: 62.4574 - val_calc_mre_K: 0.3817\n", "Epoch 1853/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.9090 - calc_mre_K: 0.3962 - val_loss: 64.9446 - val_calc_mre_K: 0.3969\n", "Epoch 1854/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.2473 - calc_mre_K: 0.3921 - val_loss: 64.9428 - val_calc_mre_K: 0.3969\n", "Epoch 1855/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.8570 - calc_mre_K: 0.3959 - val_loss: 71.6724 - val_calc_mre_K: 0.4380\n", "Epoch 1856/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.7826 - calc_mre_K: 0.3954 - val_loss: 65.5982 - val_calc_mre_K: 0.4009\n", "Epoch 1857/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.5061 - calc_mre_K: 0.3937 - val_loss: 64.3457 - val_calc_mre_K: 0.3933\n", "Epoch 1858/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.1273 - calc_mre_K: 0.3914 - val_loss: 76.0073 - val_calc_mre_K: 0.4645\n", "Epoch 1859/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.9185 - calc_mre_K: 0.3962 - val_loss: 64.0024 - val_calc_mre_K: 0.3912\n", "Epoch 1860/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.7060 - calc_mre_K: 0.3949 - val_loss: 63.7030 - val_calc_mre_K: 0.3893\n", "Epoch 1861/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 64.0628 - calc_mre_K: 0.3910 - val_loss: 63.5231 - val_calc_mre_K: 0.3882\n", "Epoch 1862/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.4637 - calc_mre_K: 0.3935 - val_loss: 66.8118 - val_calc_mre_K: 0.4083\n", "Epoch 1863/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 65.3185 - calc_mre_K: 0.3987 - val_loss: 64.1252 - val_calc_mre_K: 0.3919\n", "Epoch 1864/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.1890 - calc_mre_K: 0.3918 - val_loss: 70.4942 - val_calc_mre_K: 0.4308\n", "Epoch 1865/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.4608 - calc_mre_K: 0.3934 - val_loss: 63.5526 - val_calc_mre_K: 0.3884\n", "Epoch 1866/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.9003 - calc_mre_K: 0.3900 - val_loss: 65.9153 - val_calc_mre_K: 0.4028\n", "Epoch 1867/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.6090 - calc_mre_K: 0.3943 - val_loss: 67.6371 - val_calc_mre_K: 0.4134\n", "Epoch 1868/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.6135 - calc_mre_K: 0.3944 - val_loss: 68.5261 - val_calc_mre_K: 0.4188\n", "Epoch 1869/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.6800 - calc_mre_K: 0.3948 - val_loss: 66.7166 - val_calc_mre_K: 0.4077\n", "Epoch 1870/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.5346 - calc_mre_K: 0.3939 - val_loss: 64.0874 - val_calc_mre_K: 0.3917\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1871/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 65.0349 - calc_mre_K: 0.3969 - val_loss: 66.6595 - val_calc_mre_K: 0.4074\n", "Epoch 1872/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.0044 - calc_mre_K: 0.3907 - val_loss: 65.0357 - val_calc_mre_K: 0.3975\n", "Epoch 1873/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.8721 - calc_mre_K: 0.3959 - val_loss: 62.4298 - val_calc_mre_K: 0.3816\n", "Epoch 1874/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.4780 - calc_mre_K: 0.3935 - val_loss: 64.7608 - val_calc_mre_K: 0.3958\n", "Epoch 1875/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 64.6010 - calc_mre_K: 0.3943 - val_loss: 63.9904 - val_calc_mre_K: 0.3911\n", "Epoch 1876/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.1494 - calc_mre_K: 0.3915 - val_loss: 63.6946 - val_calc_mre_K: 0.3893\n", "Epoch 1877/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.0429 - calc_mre_K: 0.3909 - val_loss: 65.9338 - val_calc_mre_K: 0.4030\n", "Epoch 1878/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.6306 - calc_mre_K: 0.3945 - val_loss: 65.4444 - val_calc_mre_K: 0.4000\n", "Epoch 1879/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.1648 - calc_mre_K: 0.3916 - val_loss: 63.0874 - val_calc_mre_K: 0.3856\n", "Epoch 1880/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.0248 - calc_mre_K: 0.3908 - val_loss: 61.8650 - val_calc_mre_K: 0.3781\n", "Epoch 1881/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.5757 - calc_mre_K: 0.3941 - val_loss: 63.6568 - val_calc_mre_K: 0.3890\n", "Epoch 1882/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.8964 - calc_mre_K: 0.3961 - val_loss: 66.7903 - val_calc_mre_K: 0.4082\n", "Epoch 1883/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.1064 - calc_mre_K: 0.3913 - val_loss: 64.4258 - val_calc_mre_K: 0.3937\n", "Epoch 1884/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.3416 - calc_mre_K: 0.3927 - val_loss: 64.4553 - val_calc_mre_K: 0.3939\n", "Epoch 1885/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 63.7956 - calc_mre_K: 0.3894 - val_loss: 66.0842 - val_calc_mre_K: 0.4039\n", "Epoch 1886/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.6032 - calc_mre_K: 0.3943 - val_loss: 62.5518 - val_calc_mre_K: 0.3823\n", "Epoch 1887/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.3226 - calc_mre_K: 0.3926 - val_loss: 63.1502 - val_calc_mre_K: 0.3859\n", "Epoch 1888/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 64.4724 - calc_mre_K: 0.3935 - val_loss: 63.0865 - val_calc_mre_K: 0.3856\n", "Epoch 1889/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.1746 - calc_mre_K: 0.3917 - val_loss: 66.4525 - val_calc_mre_K: 0.4061\n", "Epoch 1890/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 64.7659 - calc_mre_K: 0.3953 - val_loss: 60.3097 - val_calc_mre_K: 0.3686\n", "Epoch 1891/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.5684 - calc_mre_K: 0.3941 - val_loss: 63.8612 - val_calc_mre_K: 0.3903\n", "Epoch 1892/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.7685 - calc_mre_K: 0.3892 - val_loss: 64.7515 - val_calc_mre_K: 0.3957\n", "Epoch 1893/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.2078 - calc_mre_K: 0.3919 - val_loss: 62.0912 - val_calc_mre_K: 0.3795\n", "Epoch 1894/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.4872 - calc_mre_K: 0.3936 - val_loss: 64.4648 - val_calc_mre_K: 0.3940\n", "Epoch 1895/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.0835 - calc_mre_K: 0.3911 - val_loss: 61.3377 - val_calc_mre_K: 0.3749\n", "Epoch 1896/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.7199 - calc_mre_K: 0.3950 - val_loss: 63.8861 - val_calc_mre_K: 0.3905\n", "Epoch 1897/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.0619 - calc_mre_K: 0.3910 - val_loss: 61.4001 - val_calc_mre_K: 0.3753\n", "Epoch 1898/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.9872 - calc_mre_K: 0.3905 - val_loss: 67.9833 - val_calc_mre_K: 0.4155\n", "Epoch 1899/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.4297 - calc_mre_K: 0.3871 - val_loss: 67.4487 - val_calc_mre_K: 0.4122\n", "Epoch 1900/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.7227 - calc_mre_K: 0.3950 - val_loss: 71.3624 - val_calc_mre_K: 0.4361\n", "Epoch 1901/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 63.7633 - calc_mre_K: 0.3892 - val_loss: 63.5104 - val_calc_mre_K: 0.3882\n", "Epoch 1902/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.0177 - calc_mre_K: 0.3907 - val_loss: 65.0549 - val_calc_mre_K: 0.3976\n", "Epoch 1903/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 64.4714 - calc_mre_K: 0.3935 - val_loss: 67.1379 - val_calc_mre_K: 0.4103\n", "Epoch 1904/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.2151 - calc_mre_K: 0.3919 - val_loss: 64.3252 - val_calc_mre_K: 0.3931\n", "Epoch 1905/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 64.0892 - calc_mre_K: 0.3912 - val_loss: 64.8876 - val_calc_mre_K: 0.3966\n", "Epoch 1906/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 64.6216 - calc_mre_K: 0.3944 - val_loss: 73.5543 - val_calc_mre_K: 0.4495\n", "Epoch 1907/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6393 - calc_mre_K: 0.3884 - val_loss: 62.1036 - val_calc_mre_K: 0.3795\n", "Epoch 1908/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.2802 - calc_mre_K: 0.3923 - val_loss: 63.3708 - val_calc_mre_K: 0.3873\n", "Epoch 1909/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.3232 - calc_mre_K: 0.3926 - val_loss: 63.8592 - val_calc_mre_K: 0.3903\n", "Epoch 1910/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.3753 - calc_mre_K: 0.3929 - val_loss: 64.4773 - val_calc_mre_K: 0.3941\n", "Epoch 1911/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.8819 - calc_mre_K: 0.3899 - val_loss: 65.2509 - val_calc_mre_K: 0.3988\n", "Epoch 1912/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.4425 - calc_mre_K: 0.3933 - val_loss: 65.6421 - val_calc_mre_K: 0.4012\n", "Epoch 1913/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.8721 - calc_mre_K: 0.3898 - val_loss: 61.9305 - val_calc_mre_K: 0.3785\n", "Epoch 1914/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.1014 - calc_mre_K: 0.3912 - val_loss: 65.4937 - val_calc_mre_K: 0.4003\n", "Epoch 1915/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.2147 - calc_mre_K: 0.3919 - val_loss: 64.4355 - val_calc_mre_K: 0.3938\n", "Epoch 1916/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.2816 - calc_mre_K: 0.3923 - val_loss: 67.5842 - val_calc_mre_K: 0.4130\n", "Epoch 1917/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6267 - calc_mre_K: 0.3883 - val_loss: 71.9383 - val_calc_mre_K: 0.4396\n", "Epoch 1918/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.5533 - calc_mre_K: 0.3940 - val_loss: 65.1768 - val_calc_mre_K: 0.3983\n", "Epoch 1919/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6644 - calc_mre_K: 0.3886 - val_loss: 65.2221 - val_calc_mre_K: 0.3986\n", "Epoch 1920/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.2533 - calc_mre_K: 0.3922 - val_loss: 66.6420 - val_calc_mre_K: 0.4073\n", "Epoch 1921/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 65.1184 - calc_mre_K: 0.3975 - val_loss: 67.9681 - val_calc_mre_K: 0.4154\n", "Epoch 1922/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.5588 - calc_mre_K: 0.3879 - val_loss: 64.3620 - val_calc_mre_K: 0.3934\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1923/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.1252 - calc_mre_K: 0.3914 - val_loss: 63.4795 - val_calc_mre_K: 0.3880\n", "Epoch 1924/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.1029 - calc_mre_K: 0.3913 - val_loss: 62.5220 - val_calc_mre_K: 0.3821\n", "Epoch 1925/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.4027 - calc_mre_K: 0.3931 - val_loss: 63.9975 - val_calc_mre_K: 0.3911\n", "Epoch 1926/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.9159 - calc_mre_K: 0.3901 - val_loss: 63.9909 - val_calc_mre_K: 0.3911\n", "Epoch 1927/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6572 - calc_mre_K: 0.3885 - val_loss: 63.3245 - val_calc_mre_K: 0.3870\n", "Epoch 1928/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.1392 - calc_mre_K: 0.3915 - val_loss: 62.9279 - val_calc_mre_K: 0.3846\n", "Epoch 1929/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.0655 - calc_mre_K: 0.3910 - val_loss: 64.0604 - val_calc_mre_K: 0.3915\n", "Epoch 1930/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6162 - calc_mre_K: 0.3883 - val_loss: 61.8338 - val_calc_mre_K: 0.3779\n", "Epoch 1931/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.5783 - calc_mre_K: 0.3881 - val_loss: 63.5903 - val_calc_mre_K: 0.3886\n", "Epoch 1932/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.1945 - calc_mre_K: 0.3918 - val_loss: 62.9165 - val_calc_mre_K: 0.3845\n", "Epoch 1933/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.6697 - calc_mre_K: 0.3886 - val_loss: 64.0039 - val_calc_mre_K: 0.3912\n", "Epoch 1934/2000\n", "48000/48000 [==============================] - 3s 58us/step - loss: 63.5458 - calc_mre_K: 0.3879 - val_loss: 63.9184 - val_calc_mre_K: 0.3906\n", "Epoch 1935/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.8403 - calc_mre_K: 0.3958 - val_loss: 73.0050 - val_calc_mre_K: 0.4462\n", "Epoch 1936/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.6775 - calc_mre_K: 0.3887 - val_loss: 61.7777 - val_calc_mre_K: 0.3776\n", "Epoch 1937/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.7305 - calc_mre_K: 0.3890 - val_loss: 65.0684 - val_calc_mre_K: 0.3977\n", "Epoch 1938/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.8781 - calc_mre_K: 0.3899 - val_loss: 64.9101 - val_calc_mre_K: 0.3967\n", "Epoch 1939/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.0247 - calc_mre_K: 0.3908 - val_loss: 61.8849 - val_calc_mre_K: 0.3782\n", "Epoch 1940/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.0117 - calc_mre_K: 0.3907 - val_loss: 62.1441 - val_calc_mre_K: 0.3798\n", "Epoch 1941/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6535 - calc_mre_K: 0.3885 - val_loss: 63.0182 - val_calc_mre_K: 0.3851\n", "Epoch 1942/2000\n", "48000/48000 [==============================] - 3s 59us/step - loss: 63.9646 - calc_mre_K: 0.3904 - val_loss: 64.0310 - val_calc_mre_K: 0.3913\n", "Epoch 1943/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6888 - calc_mre_K: 0.3887 - val_loss: 63.9972 - val_calc_mre_K: 0.3911\n", "Epoch 1944/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 64.3158 - calc_mre_K: 0.3926 - val_loss: 62.9713 - val_calc_mre_K: 0.3849\n", "Epoch 1945/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.3667 - calc_mre_K: 0.3868 - val_loss: 63.1366 - val_calc_mre_K: 0.3859\n", "Epoch 1946/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.9726 - calc_mre_K: 0.3905 - val_loss: 62.0156 - val_calc_mre_K: 0.3790\n", "Epoch 1947/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.8254 - calc_mre_K: 0.3896 - val_loss: 68.0797 - val_calc_mre_K: 0.4160\n", "Epoch 1948/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.7162 - calc_mre_K: 0.3889 - val_loss: 68.4876 - val_calc_mre_K: 0.4186\n", "Epoch 1949/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.5240 - calc_mre_K: 0.3877 - val_loss: 68.6185 - val_calc_mre_K: 0.4194\n", "Epoch 1950/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.5511 - calc_mre_K: 0.3940 - val_loss: 72.6270 - val_calc_mre_K: 0.4439\n", "Epoch 1951/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.9962 - calc_mre_K: 0.3906 - val_loss: 60.7515 - val_calc_mre_K: 0.3713\n", "Epoch 1952/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6863 - calc_mre_K: 0.3887 - val_loss: 64.4274 - val_calc_mre_K: 0.3938\n", "Epoch 1953/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.5372 - calc_mre_K: 0.3878 - val_loss: 61.6260 - val_calc_mre_K: 0.3766\n", "Epoch 1954/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.8161 - calc_mre_K: 0.3895 - val_loss: 62.5442 - val_calc_mre_K: 0.3822\n", "Epoch 1955/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6600 - calc_mre_K: 0.3885 - val_loss: 60.8886 - val_calc_mre_K: 0.3721\n", "Epoch 1956/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.5924 - calc_mre_K: 0.3942 - val_loss: 63.1771 - val_calc_mre_K: 0.3861\n", "Epoch 1957/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6375 - calc_mre_K: 0.3884 - val_loss: 64.1949 - val_calc_mre_K: 0.3923\n", "Epoch 1958/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.2004 - calc_mre_K: 0.3857 - val_loss: 67.0976 - val_calc_mre_K: 0.4101\n", "Epoch 1959/2000\n", "48000/48000 [==============================] - 3s 60us/step - loss: 63.8556 - calc_mre_K: 0.3897 - val_loss: 64.9651 - val_calc_mre_K: 0.3970\n", "Epoch 1960/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.8946 - calc_mre_K: 0.3900 - val_loss: 72.2572 - val_calc_mre_K: 0.4416\n", "Epoch 1961/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.5870 - calc_mre_K: 0.3881 - val_loss: 68.8547 - val_calc_mre_K: 0.4208\n", "Epoch 1962/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.8978 - calc_mre_K: 0.3900 - val_loss: 67.4090 - val_calc_mre_K: 0.4120\n", "Epoch 1963/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.1442 - calc_mre_K: 0.3915 - val_loss: 65.1297 - val_calc_mre_K: 0.3980\n", "Epoch 1964/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.8110 - calc_mre_K: 0.3895 - val_loss: 64.0033 - val_calc_mre_K: 0.3912\n", "Epoch 1965/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.6913 - calc_mre_K: 0.3887 - val_loss: 65.1381 - val_calc_mre_K: 0.3981\n", "Epoch 1966/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.9408 - calc_mre_K: 0.3903 - val_loss: 62.4823 - val_calc_mre_K: 0.3819\n", "Epoch 1967/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.7124 - calc_mre_K: 0.3889 - val_loss: 65.0622 - val_calc_mre_K: 0.3976\n", "Epoch 1968/2000\n", "48000/48000 [==============================] - 2s 51us/step - loss: 63.4077 - calc_mre_K: 0.3870 - val_loss: 61.3479 - val_calc_mre_K: 0.3749\n", "Epoch 1969/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 64.1583 - calc_mre_K: 0.3916 - val_loss: 63.4484 - val_calc_mre_K: 0.3878\n", "Epoch 1970/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.9027 - calc_mre_K: 0.3900 - val_loss: 63.3211 - val_calc_mre_K: 0.3870\n", "Epoch 1971/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.4103 - calc_mre_K: 0.3870 - val_loss: 67.9179 - val_calc_mre_K: 0.4151\n", "Epoch 1972/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.7330 - calc_mre_K: 0.3890 - val_loss: 62.1828 - val_calc_mre_K: 0.3800\n", "Epoch 1973/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.3732 - calc_mre_K: 0.3868 - val_loss: 62.5253 - val_calc_mre_K: 0.3821\n", "Epoch 1974/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 64.1914 - calc_mre_K: 0.3918 - val_loss: 65.9254 - val_calc_mre_K: 0.4029\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1975/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 63.0170 - calc_mre_K: 0.3846 - val_loss: 65.1875 - val_calc_mre_K: 0.3984\n", "Epoch 1976/2000\n", "48000/48000 [==============================] - 2s 52us/step - loss: 63.4912 - calc_mre_K: 0.3875 - val_loss: 65.3732 - val_calc_mre_K: 0.3995\n", "Epoch 1977/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.1003 - calc_mre_K: 0.3912 - val_loss: 68.5242 - val_calc_mre_K: 0.4188\n", "Epoch 1978/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.8897 - calc_mre_K: 0.3900 - val_loss: 62.0328 - val_calc_mre_K: 0.3791\n", "Epoch 1979/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.6790 - calc_mre_K: 0.3887 - val_loss: 63.9140 - val_calc_mre_K: 0.3906\n", "Epoch 1980/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.5959 - calc_mre_K: 0.3882 - val_loss: 65.3871 - val_calc_mre_K: 0.3996\n", "Epoch 1981/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.5618 - calc_mre_K: 0.3880 - val_loss: 63.6319 - val_calc_mre_K: 0.3889\n", "Epoch 1982/2000\n", "48000/48000 [==============================] - 2s 50us/step - loss: 63.0724 - calc_mre_K: 0.3850 - val_loss: 69.9363 - val_calc_mre_K: 0.4274\n", "Epoch 1983/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.7326 - calc_mre_K: 0.3890 - val_loss: 66.0826 - val_calc_mre_K: 0.4039\n", "Epoch 1984/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 64.4469 - calc_mre_K: 0.3934 - val_loss: 63.6124 - val_calc_mre_K: 0.3888\n", "Epoch 1985/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 63.1444 - calc_mre_K: 0.3854 - val_loss: 65.4492 - val_calc_mre_K: 0.4000\n", "Epoch 1986/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.1599 - calc_mre_K: 0.3855 - val_loss: 61.6598 - val_calc_mre_K: 0.3768\n", "Epoch 1987/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 63.6448 - calc_mre_K: 0.3885 - val_loss: 64.3814 - val_calc_mre_K: 0.3935\n", "Epoch 1988/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.9509 - calc_mre_K: 0.3903 - val_loss: 67.7846 - val_calc_mre_K: 0.4143\n", "Epoch 1989/2000\n", "48000/48000 [==============================] - 3s 57us/step - loss: 63.5702 - calc_mre_K: 0.3880 - val_loss: 62.0574 - val_calc_mre_K: 0.3793\n", "Epoch 1990/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.2351 - calc_mre_K: 0.3860 - val_loss: 61.6198 - val_calc_mre_K: 0.3766\n", "Epoch 1991/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 63.4371 - calc_mre_K: 0.3872 - val_loss: 68.3455 - val_calc_mre_K: 0.4177\n", "Epoch 1992/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 64.2125 - calc_mre_K: 0.3919 - val_loss: 63.4477 - val_calc_mre_K: 0.3878\n", "Epoch 1993/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.4842 - calc_mre_K: 0.3875 - val_loss: 62.1046 - val_calc_mre_K: 0.3796\n", "Epoch 1994/2000\n", "48000/48000 [==============================] - 3s 56us/step - loss: 62.7564 - calc_mre_K: 0.3830 - val_loss: 67.8222 - val_calc_mre_K: 0.4145\n", "Epoch 1995/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 64.4487 - calc_mre_K: 0.3934 - val_loss: 62.2246 - val_calc_mre_K: 0.3803\n", "Epoch 1996/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.1439 - calc_mre_K: 0.3854 - val_loss: 64.6797 - val_calc_mre_K: 0.3953\n", "Epoch 1997/2000\n", "48000/48000 [==============================] - 3s 52us/step - loss: 63.4172 - calc_mre_K: 0.3871 - val_loss: 64.6995 - val_calc_mre_K: 0.3954\n", "Epoch 1998/2000\n", "48000/48000 [==============================] - 3s 54us/step - loss: 63.5613 - calc_mre_K: 0.3879 - val_loss: 68.1855 - val_calc_mre_K: 0.4167\n", "Epoch 1999/2000\n", "48000/48000 [==============================] - 3s 53us/step - loss: 63.6430 - calc_mre_K: 0.3884 - val_loss: 70.4502 - val_calc_mre_K: 0.4306\n", "Epoch 2000/2000\n", "48000/48000 [==============================] - 3s 55us/step - loss: 63.4356 - calc_mre_K: 0.3872 - val_loss: 60.9374 - val_calc_mre_K: 0.3724\n" ] } ], "source": [ "%autoreload\n", "# import warnings\n", "# warnings.filterwarnings('ignore')\n", "\n", "\n", "# model = resnetb()\n", "# #model = conv1d_lkyrelu()\n", "# #model = conv1d_model_bnorm()\n", "# #model = conv1d_model(1)\n", "\n", "# #model = naiveploss_model()\n", "# model.summary()\n", "\n", " \n", "# from IPython.display import SVG\n", "# from keras.utils.vis_utils import model_to_dot\n", "\n", "# #SVG(model_to_dot(model).create(prog='dot', format='svg'))\n", " \n", "\n", "\n", " \n", " \n", "# import scnets as scn\n", "\n", "# model = scn.resnet(in_size=8, \n", "# out_size=256,\n", "# num_units=3,\n", "# red_dim=16,\n", "# batch_size=64,\n", "# ker_size=3)\n", "\n", "# model = scn.conv1dmodel(in_size=8, \n", "# out_size=256,\n", "# batch_size=64,\n", "# c1_nf=64,\n", "# clayers=4,\n", "# ker_size=5)\n", "\n", "model = scn.convprel(in_size=8, \n", " out_size=256,\n", " batch_size=64,\n", " c1_nf=64,\n", " clayers=4,\n", " ker_size=3)\n", "\n", "# model = scn.fullycon(in_size=8, \n", "# out_size=256, \n", "# batch_size=64,\n", "# N_hidden=2, \n", "# N_neurons=256, \n", "# N_gpus=1)\n", "\n", "\n", "\n", "\n", "\n", "# # from keras import optimizers\n", "# # sgd = optimizers.SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)\n", "# #model.compile(loss=naive_percent_loss, optimizer='nadam', metrics=[calc_mre_K])\n", "\n", "# from IPython.display import SVG\n", "# from keras.utils.vis_utils import model_to_dot\n", "# from keras.utils.vis_utils import plot_model\n", "\n", "# SVG(model_to_dot(model, show_shapes=False, show_layer_names=False).create(prog='dot', format='svg'))\n", "\n", "\n", "# #plot_model(model, show_shapes=False, show_layer_names=False, to_file='model_resnet.svg')\n", "\n", "\n", "\n", "model.summary() \n", " \n", " \n", "x_t, x_v, y_t, y_v = train_test_split(x_train, y_train, test_size=0.2, random_state=42)\n", "# model = naiveploss_mgpu_model()\n", "# model.summary() \n", "history = model.fit(x_t, y_t,\n", " batch_size=64,\n", " epochs=2000, \n", " verbose=1,\n", " validation_data=(x_v, y_v))\n", "\n", "\n", "\n", "\n", "\n", "# y_pred = model.predict(x_test)\n", "# print(calc_mre(y_test, y_pred))\n", "\n" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "ExecuteTime": { "end_time": "2018-09-30T16:23:59.012349Z", "start_time": "2018-09-30T16:23:58.711234Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/hegder/anaconda3/envs/dp2/lib/python3.5/site-packages/keras/backend/mxnet_backend.py:89: UserWarning: MXNet Backend performs best with `channels_first` format. Using `channels_last` will significantly reduce performance due to the Transpose operations. For performance improvement, please use this API`keras.utils.to_channels_first(x_input)`to transform `channels_last` data to `channels_first` format and also please change the `image_data_format` in `keras.json` to `channels_first`.Note: `x_input` is a Numpy tensor or a list of Numpy tensorRefer to: https://github.com/awslabs/keras-apache-mxnet/tree/master/docs/mxnet_backend/performance_guide.md\n", " train_symbol = func(*args, **kwargs)\n", "/home/hegder/anaconda3/envs/dp2/lib/python3.5/site-packages/keras/backend/mxnet_backend.py:92: UserWarning: MXNet Backend performs best with `channels_first` format. Using `channels_last` will significantly reduce performance due to the Transpose operations. For performance improvement, please use this API`keras.utils.to_channels_first(x_input)`to transform `channels_last` data to `channels_first` format and also please change the `image_data_format` in `keras.json` to `channels_first`.Note: `x_input` is a Numpy tensor or a list of Numpy tensorRefer to: https://github.com/awslabs/keras-apache-mxnet/tree/master/docs/mxnet_backend/performance_guide.md\n", " test_symbol = func(*args, **kwargs)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "__________________________________________________________________________________________________\n", "Layer (type) Output Shape Param # Connected to \n", "==================================================================================================\n", "input_1 (InputLayer) (None, 8) 0 \n", "__________________________________________________________________________________________________\n", "dense_1 (Dense) (None, 256) 2304 input_1[0][0] \n", "__________________________________________________________________________________________________\n", "p_re_lu_1 (PReLU) (None, 256) 256 dense_1[0][0] \n", "__________________________________________________________________________________________________\n", "reshape_1 (Reshape) (None, 8, 32) 0 p_re_lu_1[0][0] \n", "__________________________________________________________________________________________________\n", "conv1d_1 (Conv1D) (None, 8, 32) 3104 reshape_1[0][0] \n", "__________________________________________________________________________________________________\n", "activation_1 (Activation) (None, 8, 32) 0 conv1d_1[0][0] \n", "__________________________________________________________________________________________________\n", "conv1d_2 (Conv1D) (None, 8, 32) 3104 activation_1[0][0] \n", "__________________________________________________________________________________________________\n", "add_1 (Add) (None, 8, 32) 0 conv1d_2[0][0] \n", " reshape_1[0][0] \n", "__________________________________________________________________________________________________\n", "conv1d_3 (Conv1D) (None, 8, 32) 3104 add_1[0][0] \n", "__________________________________________________________________________________________________\n", "activation_2 (Activation) (None, 8, 32) 0 conv1d_3[0][0] \n", "__________________________________________________________________________________________________\n", "conv1d_4 (Conv1D) (None, 8, 32) 3104 activation_2[0][0] \n", "__________________________________________________________________________________________________\n", "add_2 (Add) (None, 8, 32) 0 conv1d_4[0][0] \n", " add_1[0][0] \n", "__________________________________________________________________________________________________\n", "flatten_1 (Flatten) (None, 256) 0 add_2[0][0] \n", "==================================================================================================\n", "Total params: 14,976\n", "Trainable params: 14,976\n", "Non-trainable params: 0\n", "__________________________________________________________________________________________________\n" ] } ], "source": [ "import scnets as scn\n", "from keras.models import load_model\n", "from scnets import relerr_loss, calc_mre_K\n", "\n", "#Creates a HDF5 file 'my_model.h5'\n", "#odel.save('res15k_model.h5')\n", "#odel = load_model('model/multi_task/try.h5', custom_objects={'loss_max': loss_max})\n", "model = load_model('res15k_model.h5', custom_objects={'relerr_loss': relerr_loss, \n", " 'calc_mre_K' :calc_mre_K})\n", "model.summary()" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "ExecuteTime": { "end_time": "2018-10-10T13:19:04.315718Z", "start_time": "2018-10-10T13:19:04.280764Z" }, "scrolled": true }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "MXNet Backend: Successfully exported the model as MXNet model!\n", "MXNet symbol file - my_mod_convprel-symbol.json\n", "MXNet params file - my_mod_convprel-0000.params\n", "\n", "\n", "Model input data_names and data_shapes are: \n", "data_names : ['/first_input6']\n", "data_shapes : [DataDesc[/first_input6,(64, 8),float32,NCHW]]\n", "\n", "\n", "Note: In the above data_shapes, the first dimension represent the batch_size used for model training. \n", "You can change the batch_size for binding the module based on your inference batch_size.\n" ] }, { "data": { "text/plain": [ "(['/first_input6'], [DataDesc[/first_input6,(64, 8),float32,NCHW]])" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Import the save_mxnet_model API\n", "from keras.models import save_mxnet_model\n", "save_mxnet_model(model=model, prefix='my_mod_convprel', epoch=0)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-30T10:30:53.029361Z", "start_time": "2018-09-30T10:30:52.979098Z" } }, "outputs": [], "source": [ "import numpy as np\n", "import mxnet as mx\n", "\n", "# Step1: Load the model in MXNet\n", "\n", "# Use the same prefix and epoch parameters we used in save_mxnet_model API.\n", "sym, arg_params, aux_params = mx.model.load_checkpoint(prefix='my_mod', epoch=0)\n", "\n", "# We use the data_names and data_shapes returned by save_mxnet_model API.\n", "mod = mx.mod.Module(symbol=sym, \n", " data_names=['/input_12'], \n", " context=mx.gpu(), \n", " label_names=None)\n", "mod.bind(for_training=False, \n", " data_shapes=[('/input_12', (1,8))], \n", " label_shapes=mod._label_shapes)\n", "mod.set_params(arg_params, aux_params, allow_missing=True)\n", "\n", "data_iter = mx.io.NDArrayIter(x_test, None, 1)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-30T10:40:42.350586Z", "start_time": "2018-09-30T10:40:42.321089Z" } }, "outputs": [], "source": [ "size = np.random.randint(30, 71, (1,8))\n", "#spec_ac = snlay.calc_spectrum(size, mats, lams)\n", "size = (size - 50.0)/20.0\n", "#size = np.expand_dims(size, axis = 0)\n", "size.shape\n", "res1 = model.predict(size)\n", "data_iter = mx.io.NDArrayIter(size, None, 1)\n", "res2 = mod.predict(data_iter)\n", "\n", "# #y_pred = result.asnumpy()\n", "# print(calc_mre(y_test, y_pred))" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-30T11:18:00.192453Z", "start_time": "2018-09-30T11:18:00.168150Z" } }, "outputs": [], "source": [ "import mxnet as mx\n", "from mxnet import nd\n", "mx.random.seed(1)\n", "\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "ExecuteTime": { "end_time": "2018-09-30T16:24:20.697887Z", "start_time": "2018-09-30T16:24:18.314292Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/hegder/anaconda3/envs/dp2/lib/python3.5/site-packages/mxnet/module/bucketing_module.py:408: UserWarning: Optimizer created manually outside Module but rescale_grad is not normalized to 1.0/batch_size/num_workers (1.0 vs. 0.03125). Is this intended?\n", " force_init=force_init)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "0.8106580427940897\n" ] } ], "source": [ "# idx = 102\n", "# print(\"Predicted - \", np.argmax(result[idx].asnumpy()))\n", "# print(\"Actual - \", y_test[idx])\n", "\n", "#result.asnumpy()[1] - y_test[1]\n", "\n", "\n", "y_pred = model.predict(x_test)\n", "print(calc_mre(y_test, y_pred))\n", "#print(calc_mre(y_test, result.asnumpy()))\n", "#history60 = history\n", "#plot_training_history(history, 32*2.56)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 4, "metadata": { "ExecuteTime": { "end_time": "2018-10-01T16:13:05.287947Z", "start_time": "2018-10-01T16:13:04.943850Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[42 52 67 48 57 66 35 64]\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "from matplotlib import gridspec\n", "import snlay as snlay\n", "#here we test edge cases where failure happens\n", "#size = np.array([70, 60, 50, 40, 30, 20, 10, 10])\n", "#size = np.array([70, 70, 70, 70, 70, 70, 70, 70])\n", "#size = np.array([30, 30, 30, 30, 30, 30, 30, 30])\n", "\n", "#size = np.array([65, 65, 65, 65, 55, 65, 35, 65])\n", "#size = np.array([65, 35, 45, 35, 45, 35, 45, 35])\n", "\n", "size = np.random.randint(30, 71, 8)\n", "mats = np.array([3, 4, 3, 4, 3, 4, 3, 4])\n", "spec_ac = snlay.calc_spectrum(size, mats, lams)\n", "\n", "print(size)\n", "\n", "\n", "size = (size - 50.0)/20.0\n", "\n", "\n", "\n", "spec = model.predict(np.expand_dims(size, axis = 0))\n", "\n", "spec = np.ravel(spec)\n", "\n", "\n", "fig1 = plt.figure(figsize=(11,3))\n", "gs = gridspec.GridSpec(1, 2, width_ratios=[8, 3]) \n", "\n", "ax = plt.subplot(gs[0])\n", "#ax = fig1.add_subplot(1,2,1)\n", "#ax.set_title('silica coated gold')\n", "ax.set_xlabel('Wavelength (nm)')\n", "ax.set_ylabel('Extinction Efficiency Qe')\n", "ax.set_ylim((0, 6))\n", "ax.set_xlim((300, 1200))\n", "plt.plot(lams, spec_ac,'b', linewidth=1, label='True')\n", "plt.plot(lams, spec, 'r--', linewidth=2, label='predicted')\n", "plt.plot(lams, 10*np.abs(spec_ac - spec)/spec_ac,'k', linewidth=1, label='10x Relative error')\n", "ax.legend(loc='best')\n", "\n", "ax2 = plt.subplot(gs[1])\n", "#fig2 = plt.figure(figsize=(3,3))\n", "#ax2 = fig1.add_subplot(1,2,2)\n", "#ax.set_title('silica coated gold')\n", "ax2.set_xlabel('Wavelength (nm)')\n", "#ax.set_ylabel('Extinction Efficiency Qe')\n", "#ax2.set_ylim((2, 6))\n", "ax2.set_xlim((300, 450))\n", "plt.plot(lams, spec_ac,'b', linewidth=1, label='True')\n", "plt.plot(lams, spec, 'r--', linewidth=2, label='predicted')\n", "#plt.plot(lams, 10*np.abs(spec_ac - spec)/spec_ac,'b', linewidth=1, label='10x Relative error')\n", "#ax.legend(loc='best')\n", "\n", "\n", "\n", "plt.tight_layout()\n", "plt.savefig('grid_figure.pdf')\n", "\n", "\n", "\n", "\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "ExecuteTime": { "end_time": "2018-10-01T16:13:32.441268Z", "start_time": "2018-10-01T16:13:32.415956Z" } }, "outputs": [ { "data": { "text/plain": [ "(8,)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "size.shape" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "ExecuteTime": { "end_time": "2018-09-30T12:48:05.013204Z", "start_time": "2018-09-30T12:48:04.790487Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "1.829695701599121\n" ] }, { "ename": "NameError", "evalue": "name 'result' is not defined", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 27\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 28\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 29\u001b[0;31m \u001b[0mresult\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", "\u001b[0;31mNameError\u001b[0m: name 'result' is not defined" ] } ], "source": [ "# loop for hundred runs \n", "import snlay as snlay\n", "import time\n", "\n", "mats = np.array([3, 4, 3, 4, 3, 4, 3, 4])\n", "#spec_ac = snlay.calc_spectrum(size, mats, lams)\n", "\n", "reps = 100\n", "\n", "\n", "start = time.time()\n", "for ind in np.arange(reps):\n", " size = np.random.randint(30, 71, 8)\n", " #size = np.random.randint(30, 71, (1,8))\n", " #spec_ac = snlay.calc_spectrum(size, mats, lams)\n", " size = (size - 50.0)/20.0\n", " size = np.expand_dims(size, axis=0)\n", " spec = model.predict(size)\n", " #data_iter = mx.io.NDArrayIter(size, None, 1)\n", " #result = mod.predict(size)\n", " #result = result.asnumpy()\n", "\n", " \n", "end = time.time()\n", "print(1000*(end - start)/reps) \n", " \n", "\n", "\n", "result\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "heading_collapsed": true }, "source": [ "### Inverse scattering " ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-29T09:59:20.270577Z", "start_time": "2018-09-29T09:59:20.206497Z" }, "hidden": true }, "outputs": [], "source": [ "# from keras.models import Model\n", "# from keras.layers import Input, Add, AveragePooling1D, MaxPooling1D, Concatenate\n", "\n", "\n", "\n", "# a = Input(shape=(1,))\n", "# first = Dense(8, input_dim=1, kernel_initializer='normal', activation='linear', \n", "# name='dummy', use_bias=False)(a)\n", "# last = model(first)\n", "# #last = Dense(256, kernel_initializer='normal')(first)\n", "# #last = Flatten()(first)\n", "\n", "# model_d = Model(inputs=a, outputs=last)\n", "# model_d.compile(loss=naive_percent_loss, optimizer='nadam', metrics=[calc_mre_K], context=['gpu(0)'])\n", "\n", "model_d = Sequential()\n", "model_d.add(Dense(8, input_dim=1, kernel_initializer='normal', activation='linear', \n", " name='dummy', use_bias=False))\n", "\n", "for layer in model.layers[1:]:\n", " model_d.add(layer)\n", "\n", "# for layer in model_d.layers[1:]:\n", "# layer.trainable = False\n", "\n", "# for ind in range(1,len(model_d.layers)):\n", "# model_d.layers[ind].set_weights(model.layers[ind-1].get_weights())\n", "\n", "model_d.compile(loss=naive_percent_loss, optimizer='adam') \n", " \n", " \n", "model_d.summary()\n", "\n", "\n", "for layer in model.layers[1:]:\n", " print(layer)\n", "\n", "\n", "\n", "\n", "\n", "\n", "# # # let us create a target spectrum first\n", "# import snlay as snlay\n", "# #size = np.array([60, 65, 65, 65, 35, 35, 35, 35])\n", "# size = np.random.randint(30,70,8)\n", "# mats = np.array([3, 4, 3, 4, 3, 4, 3, 4])\n", "# target = snlay.calc_spectrum(size, mats, lams)\n", "\n", "# print(size)\n", "\n", "# # #do the training here\n", "# xd_t = np.ones((1,1))\n", "# yd_t = target.reshape(1,250)\n", "\n", "\n", "\n", "\n", "\n", "# history = model_d.fit(xd_t, yd_t,\n", "# batch_size=1,\n", "# epochs=5000, \n", "# verbose=0)\n", "\n", "# # #here is the final result\n", "# size_out = model_d.get_layer('dummy')\n", "# wts = size_out.get_weights()\n", "# wts = np.array(wts).ravel()\n", "# size_res= 0.5*wts*(size_max - size_min) + size_av\n", "# size_res_rounded = np.round(size_res)\n", "\n", "# spec_zer = model_d.predict(xd_t).ravel()\n", "# achieved = snlay.calc_spectrum(size_res_rounded, mats, lams)\n", "\n", "\n", "# fig1 = plt.figure(figsize=(22,5))\n", "# ax = fig1.add_subplot(1,1,1)\n", "# #plt.plot(lams, spec_zer, label='new model')\n", "# plt.plot(lams, target, linewidth=2, label='target')\n", "# plt.plot(lams, achieved, '--', linewidth=3, label='achieved')\n", "# plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)\n", "\n", "# print(size_res_rounded)\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-26T10:28:37.316134Z", "start_time": "2018-09-26T10:27:22.339834Z" }, "hidden": true }, "outputs": [], "source": [ "from keras.models import Model\n", "from keras.layers import Input, Add, AveragePooling1D, MaxPooling1D, Concatenate\n", "\n", "a = Input(shape=(8,))\n", "first = Dense(256, kernel_initializer='normal')(a)\n", "#first = Dense(128, kernel_initializer='normal')(first)\n", "#first = BatchNormalization()(first)\n", "first= Activation('relu')(first)\n", "\n", "first = Reshape((256,1))(first)\n", "#first = UpSampling1D(size = 2)(first)\n", "first = Conv1D(filters=32, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal', activation='relu')(first)\n", "#first = UpSampling1D(size = 2)(first)\n", "\n", "\n", "\n", "\n", "# first_1 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", "# kernel_initializer='normal')(first)\n", "# first_1 = Activation('relu')(first_1)\n", "# first_1 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", "# kernel_initializer='normal')(first_1)\n", "first_2 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "#first_2 = Activation('relu')(first_2)\n", "first_2 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", " kernel_initializer='normal')(first_2)\n", "first_3 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "first_3 = Activation('relu')(first_3)\n", "first_3 = Conv1D(filters=8, kernel_size=5, strides=1, padding='same', \n", " kernel_initializer='normal')(first_3)\n", "first_4 = Conv1D(filters=16, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "list_of_tensors = [first_2, first_3, first_4]\n", "conc = Concatenate()(list_of_tensors)\n", "first = Add()([first, conc])\n", "first= Activation('relu')(first)\n", "\n", "\n", "\n", "# first_1 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", "# kernel_initializer='normal')(first)\n", "# first_1 = Activation('relu')(first_1)\n", "# first_1 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", "# kernel_initializer='normal')(first_1)\n", "first_2 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "#first_2 = Activation('relu')(first_2)\n", "first_2 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", " kernel_initializer='normal')(first_2)\n", "first_3 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "first_3 = Activation('relu')(first_3)\n", "first_3 = Conv1D(filters=8, kernel_size=5, strides=1, padding='same', \n", " kernel_initializer='normal')(first_3)\n", "first_4 = Conv1D(filters=16, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "list_of_tensors = [first_2, first_3, first_4]\n", "conc = Concatenate()(list_of_tensors)\n", "first = Add()([first, conc])\n", "first= Activation('relu')(first)\n", "\n", "\n", "\n", "\n", "# first_1 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", "# kernel_initializer='normal')(first)\n", "# first_1 = Activation('relu')(first_1)\n", "# first_1 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", "# kernel_initializer='normal')(first_1)\n", "first_2 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "#first_2 = Activation('relu')(first_2)\n", "first_2 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", " kernel_initializer='normal')(first_2)\n", "first_3 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "first_3 = Activation('relu')(first_3)\n", "first_3 = Conv1D(filters=8, kernel_size=5, strides=1, padding='same', \n", " kernel_initializer='normal')(first_3)\n", "first_4 = Conv1D(filters=16, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "list_of_tensors = [first_2, first_3, first_4]\n", "conc = Concatenate()(list_of_tensors)\n", "first = Add()([first, conc])\n", "first= Activation('relu')(first)\n", "\n", "# first_1 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", "# kernel_initializer='normal')(first)\n", "# first_1 = Activation('relu')(first_1)\n", "# first_1 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", "# kernel_initializer='normal')(first_1)\n", "first_2 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "#first_2 = Activation('relu')(first_2)\n", "first_2 = Conv1D(filters=8, kernel_size=3, strides=1, padding='same', \n", " kernel_initializer='normal')(first_2)\n", "first_3 = Conv1D(filters=4, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "first_3 = Activation('relu')(first_3)\n", "first_3 = Conv1D(filters=8, kernel_size=5, strides=1, padding='same', \n", " kernel_initializer='normal')(first_3)\n", "first_4 = Conv1D(filters=16, kernel_size=1, strides=1, padding='same', \n", " kernel_initializer='normal')(first)\n", "list_of_tensors = [first_2, first_3, first_4]\n", "conc = Concatenate()(list_of_tensors)\n", "first = Add()([first, conc])\n", "first= Activation('relu')(first)\n", "\n", "\n", "\n", "\n", "\n", "first = Reshape((32,256))(first)\n", "first = MaxPooling1D(pool_size=32, strides=None, padding='same')(first)\n", "last = Flatten()(first)\n", "\n", "model = Model(inputs=a, outputs=last)\n", "model.compile(loss=naive_percent_loss, optimizer='nadam', metrics=[calc_mre_K], context=['gpu(0)'])\n", "\n", "#model.summary()\n", "\n", "from IPython.display import SVG\n", "from keras.utils.vis_utils import model_to_dot\n", "from keras.utils.vis_utils import plot_model\n", "\n", "#SVG(model_to_dot(model, show_shapes=True, show_layer_names=False).create(prog='dot', format='svg'))\n", "\n", "\n", "plot_model(model, show_shapes=False, show_layer_names=False, to_file='model.png')\n", "#plot_model(model, to_file='model.png', )\n", "\n", "x_t, x_v, y_t, y_v = train_test_split(x_train, y_train, test_size=0.2, random_state=42)\n", "# model = naiveploss_mgpu_model()\n", "# model.summary() \n", "history = model.fit(x_train, y_train,\n", " batch_size=64,\n", " epochs=2000, \n", " verbose=1,\n", " validation_data=(x_test, y_test))\n", "\n", "\n", "\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "heading_collapsed": true }, "source": [ "### Model shipment" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-06T10:51:07.997455Z", "start_time": "2018-09-06T10:49:19.326Z" }, "hidden": true }, "outputs": [], "source": [ "\n", " \n", "from keras.models import load_model\n", "\n", "#Creates a HDF5 file 'my_model.h5'\n", "model.save('my_model.h5')\n", "\n", "# Deletes the existing model\n", "#del model \n", "\n", "# Returns a compiled model identical to the previous one\n", "#model = load_model('my_model.h5')\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-08T09:40:51.323101Z", "start_time": "2018-09-08T09:40:50.155098Z" }, "hidden": true }, "outputs": [], "source": [ "\n", "from keras.models import Model\n", "from keras.layers import Input, Add, Lambda, Dense\n", "import numpy as np\n", "\n", "\n", "def dropper(x):\n", " ms = 4\n", " #print(x.shape)\n", " \n", " \n", " \n", " return x**2\n", "# msk = np.array([1,1,1,1,0,0,0,0])\n", "\n", "\n", "a = Input(shape=(1,))\n", "b = Dense(8, input_dim=1, kernel_initializer='normal', activation='linear', \n", " name='dummy', use_bias=False)(a)\n", "b = Lambda(dropper)(b)\n", "b = Dense(256)(b)\n", "# #b = Lambda()\n", "# a = Lambda(dropper)(a)\n", "# \n", "\n", "model = Model(inputs=a, outputs=b)\n", "\n", "# model = Sequential()\n", "# model.add(Dense(256, input_dim=8))\n", "# #model.add(Lambda(lambda x: x**2))\n", "# model.add(Lambda(dropper))\n", "# model.add(Dense(256))\n", "\n", "\n", "\n", "model.compile(loss=naive_percent_loss, optimizer='adam', metrics=[calc_mre_K])\n", "#model.summary()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-28T10:32:16.469180Z", "start_time": "2018-09-28T10:32:16.452138Z" }, "hidden": true }, "outputs": [], "source": [ "2 +2 " ] }, { "cell_type": "code", "execution_count": null, "metadata": { "ExecuteTime": { "end_time": "2018-09-28T10:32:36.778144Z", "start_time": "2018-09-28T10:32:36.682162Z" }, "hidden": true }, "outputs": [], "source": [ "import matplotlib.pyplot as plt" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "hidden": true }, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.5.4" } }, "nbformat": 4, "nbformat_minor": 2 }