{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": { "ExecuteTime": { "end_time": "2018-12-23T19:21:39.142476Z", "start_time": "2018-12-23T19:21:38.537604Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Dataset has been loaded\n", "x-train (180000, 16)\n", "x-test (20000, 16)\n", "y-train (180000, 128)\n", "y-test (20000, 128)\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/s16_d_siti_2.h5','r')\n", "X = h5f['sizes'][:]\n", "Y = h5f['spectrum'][:]\n", "\n", "#get the ranges of the loaded data\n", "num_layers = X.shape[1]\n", "num_lpoints = Y.shape[1]\n", "\n", "\n", "#this information is not given in the dataset\n", "lam_min = 400\n", "lam_max = 800\n", "lams = np.linspace(lam_min, lam_max, num_lpoints, endpoint=True)\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=0.1, random_state=42)\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", "\n", "\n" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "ExecuteTime": { "end_time": "2018-12-23T19:22:11.284971Z", "start_time": "2018-12-23T19:22:11.172903Z" } }, "outputs": [ { "data": { "text/plain": [ "(0, 100)" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD/CAYAAAD/qh1PAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3XmUnHWd7/H3r6r3bJ2ks290QpYnCQphcRS4WIi4oIWIonBlmQHvnDPc0XMYZu6M3HtPXI7O1YPOweU4FxT1jiPIpiWrKMUSBEQCIskD2Tqd9JZ0J51O711d9bt/PE+HopNeaumu7qc+r3PqVPXzq+epX39T+dSvn+f3PGWstYiISHCFCt0BERGZWAp6EZGAU9CLiAScgl5EJOAU9CIiAaegFxEJOAW9iEjAlYznSbG4ezNwPfAu4MVoxHl/WttM4IdAFOgHfgT8SzTi2PG0i4jIxBpX0APNwL8C5wLvHdZ2B7AAWAVUA08CTf7y8bSLiMgEGteum2jEeTAacR4EDqUvj8XdKuAa4LZoxGmPRpw64FvAjeNpFxGRiTfeEf1I1gFlwGtpy7YDm2JxNzxWezTiJIdv8Jt3PmGA5cDxHPsmIlJMZgMN//T5D520WzzXoJ8F9EQjzmDasmNAGKgcR3vXKba5HDiQY79ERIrRSuDg8IW5Bn0nUBWLuyVpYV4NJIHecbSfynGAxx76CYODiYw6Y0Ihlq5aT1P9W9hUKsNfpfioXplRvTKjemUml3qVlJTykStugBH2hOQa9LuAAeDdwCv+srOAndGIk4zF3VHbR9twYqCfwcRARp0xoRDJZJJEf5/eWOOgemVG9cqM6pWZXOo11vPHO72yxH9uCRCKxd0KIBWNOD2xuPufwNdicfdqvNH6rcD3AMZqFxGRiTfeE6b+J96ulm8BF/qPf+u3fRE4grdf/U/Ag8B309Ydq11ERCbQuEb00YizFdg6Qlsn8LlR1h21XUREJpYugSAiEnAKehGRgFPQi4gEnIJeRCTgFPQiIgGnoBcRCTgFvYhIwCnoRUQCTkEvIhJwCnoRkYBT0IuIBJyCXkQk4BT0IiIBp6AXEQk4Bb2ISMAp6EVEAk5BLyIScAp6EZGAU9CLiAScgl5EJOAU9CIiAaegFxEJOAW9iEjAKehFRAJOQS8iEnAKehGRgFPQi4gEnIJeRCTgFPQiIgGnoBcRCTgFvYhIwCnoRUQCTkEvIhJwCnoRkYArycdGYnF3KfBd4CLAANuAm6MRpyEWd0uA24Fr8T5Y7vfb+vPx2iIiMrp8jeh/AJQCtcAKoBu402/7Et4HwGZgLbAJ+EaeXldERMaQlxE9sBr4P9GI0wkQi7s/B/7db7sJuCUacZr8tq3APbG4e2s04qRG2qAJhTChzD6Hhp6f6XrFSvXKjOqVGdUrM7nUa6x18hX03wauisXdR4Ak3m6aR2JxtxpvhP9q2nO3A0PL60fa4LJah2QymVVnltVuzGq9YqV6ZUb1yozqlZls6hUOh0dtz1fQPw/cCBwFLPBn4APALL+9I+25x/z7WYyisc5lMDGQUSdMKMSy2o001u3Epkb8Y0F8qldmVK/MqF6ZyaVeJaVlbDnngpHbc+1cLO6GgCeBXwIf8hdvBR4DPuz/PAdo8x9X+/edo23XplJZvzlyWbcYqV6ZUb0yo3plJpt6jfX8fOw8mwesAu6IRpyeaMTpwZuB8x68D5KDwJlpzz8Lb1R/MA+vLSIiY8g56KMRpw3YA9wci7sVsbhbAfw90OC33QXcFou7S2NxdwHeaP/u0Q7EiohI/uTrcPjlwBagEWgCzgGiftvX8ebV78D7QNiJN+VSREQmQV4OxkYjzk7e3j8/vG0Q+IJ/ExGRSaYJriIiAaegFxEJOAW9iEjAKehFRAJOQS8iEnAKehGRgFPQi4gEnIJeRCTgFPQiIgGnoBcRCTgFvYhIwCnoRUQCTkEvIhJwCnoRkYBT0IuIBJyCXkQk4BT0IiIBp6AXEQk4Bb2ISMAp6EVEAk5BLyIScAp6EZGAU9CLiAScgl5EJOAU9CIiAaegFxEJOAW9iEjAKehFRAJOQS8iEnAKehGRgFPQi4gEnIJeRCTgFPQiIgFXkq8NxeLux4EvA+uATuDb0YjzrVjcLQFuB67F+2C5H7g5GnH68/XaIiIysryM6GNx98PAD4F/BKqB9cBjfvOXgIuAzcBaYBPwjXy8roiIjC1fI/qvAl+LRpzf+z8fB97wH98E3BKNOE0Asbi7FbgnFndvjUacVJ5eX0RERpBz0Mfi7gzgbLzw3gnMB/4AfBEv8FcAr6atsh1v1L8CqB9puyYUwoQy+4Nj6PmZrlesVK/MqF6ZUb0yk0u9xlonHyP6uYABrgc+ChwGvgM8AHzSf05H2vOP+fezRtvoslqHZDKZVYeW1W7Mar1ipXplRvXKjOqVmWzqFQ6HR23PR9B3+vd3RCPOfoBY3L0NaAWGknoO0OY/rh623ik11rkMJgYy6ogJhVhWu5HGup3YlPYKjUX1yozqlRnVKzO51KuktIwt51wwcnuunYtGnI5Y3K0HbNri9McHgTOBvf7PZ+GN6g+Otl2bSmX95shl3WKkemVG9cqM6pWZbOo11vPzdTD2h8AXY3H3t3gj+a8Cf4pGnKZY3L0LuC0Wd18AEsBW4G4diBURmRz5Cvpv4u2r3443ZXMbcKXf9nWgBtjht92HN+VSREQmQV6C3h+d/w//NrxtEPiCfxMRkUmmeU8iIgGnoBcRCTgFvYhIwCnoRUQCTkEvIhJwCnoRkYBT0IuIBJyCXkQk4BT0IiIBp6AXEQm4vH1nrIzMlldD5SIoqYTBHug8iEn2FrpbIlIkFPQTyM45HbvkfTBzBQx0eiFfOhNKKrHHdmMansL0Hy10N0Uk4BT0E8CGK7CnXQZzTofDL2P2xTAD3hdrWYAZS7GL34vd9N+gMQ6HXsIUtMciEmQK+jyzFfOxp18FyT7MjjtPGrEbgO4mzN4HsHPWYldfDhU1UP8o5h3f1yIikh86GJtHtmI+dv3noLsR8+bPxtwtYzp2Y9yfQvXp2FUfUcyLyIQomhG9t8tkGcyuxZowpqcFju/FpAbzs/3yedh1n4OOvZj9D497V4zpa4Vdv8Cu/xw20YVpejYv/RERGVIUQW/LZnv7zGeugq6DkEpgF24Bm4KGp+HIn3PaR27L53oj+c46zP5HMt6W6T0Me+/Hrr0G292M6didQ29ERN4p8EFvy+d5Idx7GPOX72MSnd5yE4KaLdgVl8CcNbD/YUxqIIvtV/shfwBT95us97ObzgPQGMfWRmHnnZiB41ltR0RkuEDvo7clVdh110BnPWb3vSdCHsDYFKb1Txj3x1AxD7vhOmzZ7My2XzbH213T1YCp+3XuB1MPvQTdDdjTLtP+ehHJm8AGvcVgV18BA8cw+0ceaZv+dsybP4OB49gNf42tWjK+7ZfN8UbyPc35CXm8GTlm/6NQtRRqzsp5eyIiEOCgZ9F7oLIGs/chjE2N+lSTGsDsuQ/ad2LXX4utXj/q823lIuyGG6CnBbNv7O1nwiQ6MQefxC6/GFsyI2/bFZHiFcigt+XzsMsuwtQ/jhnsHtc6Bkvo4JOYhqewq68gteKD2FDpO7eLwS44B+tcD8fexOx9MK8hf8KR16G3Fbv84vxvW0SKTuAOxlogtfwS6NiDOfZWxuub1j9Bbwt21WXY+Wdgj7qY/iPY0pkwdwOEKzB1D2Pad+a/80N9ADjwOHbjjdi21zBdByfstUQk+AI3ou8tWQizVmEO/i7rbZiuBszOOzH1j0K4DDt3A5TPxRx6GfP69yY05E/0ofcwtG7HLv+ADsyKSE4CNaK3JkR7xSbMoZcwAx05bcvYFLS/iWl/M0+9y6IPTduwZ/wdVK+HLP46mQ4sQKgETBhSSYzNzwlsIvK2YAX9gnO9s14PvVDoruSFGezGHnoRuzwCx3YF4lo4tmK+d3byzJVQWQPl1ZB2LMQO9kGiE3oOYbqb4Pg+TN+RAvZYZPoLVNATLmdu307aU4kARKLHtLyEXXA21JwJba8WujtZsaEyqHkXtubdULUYeg555zZ07IH+dhjs9c5SNmEonQHlc7BVS7Dz3wUrL8X2tmGO7oS2VzGJrkL/OiLTTqCCPtT8LDOqNtNe6I7kkUkNQNM27NIL4egbmFSi0F0aNxsqhYXnYhe/B5L9mNbXYPcv33Hi2kn6WqETDH/2tlE6C6rXYWvOhKUXYNtdTNM2TF/bJP0WItNfoII+sNq2w6LzvFvz84XuzbjYOWuxKz8ENukdGD/yRla7nkyiE1pf8W4nruP/eezRHZjGp3WpCJFxUNBPA8amoOkZ7MoPw+FXMMm+QndpRDZc7vVz7gZM8/PQ8gLGJnPe7juu41+1BLs8gt30t9D8HBz648SczyASEIGbXhlYR3d4l2lY/FeF7smIbNVi7MYboXIBZuddmOZteQn54UxPM2bXf2L2P4xdeB7W+RtsRU3eX0ckKBT004QBTNMz3j7vkqpCd+ckdu4G7IbrvOvxu3dP+EwZA5h2F7Pj36HnkHdy2cJzA3MQXiSfFPTTybFd0NeGXfy+QvfkBAvefvPaT2AOPEnowBMTMoofiUn2E9r/G0zdr7FL/wt29Se9WT4icoKCfhoxgGl8Bhae7c1GKTAL3sXXllyA2XMvpoDTP037m29fctq5AVs+r2B9EZlq8nowNhZ3K4G/ADXRiFPtLysBbgeuxftguR+4ORpx+vP52kXj+D7obsIuOR9z4PGCdcOCd9B13kbMrp97JzcVmOlvhzd/6l2nyLke9tyv6wSJkP8R/VeAhmHLvgRcBGwG1gKbgG/k+XWLhjeqfxpqzsSWVResH3bZ+72Qf+s/pkTIDzGpBKbuV3B4O3bdNdi5GwvdJZGCy9uIPhZ3twAfBW4B7k1rugm4JRpxmvznbQXuicXdW6MRZ8Q5cSYUwoQy+xwaen6m6003pqeRZNcBWHYhpv6R7LeTZb1SC8+DhecS2nMPpr8Npli9DUDLc6QSHd5XM5ZWEWrbnvt2i+T9lS+qV2ZyqddY6+Ql6P3dM3cCNw9bXg2sANJ33m4HhpbXj7TNZbUOyWR2B/WW1QZ/FNcfaqBl3vksKjtCaSq3ywJkUq/O0hUcrXwXC3v+SOXiucDcnF57YqXo6d1O64pLmLVwJXP69+Rlq8Xw/son1Ssz2dQrHA6P2p6vEf0/AK9HI87Tsbj7/rTlQ0cM0y8leWxY2yk11rkMJjL7sm4TCrGsdiONdTuxqSI4gWb1IprsYsJ1v8pq9UzrZavXk6o9A7M/xpF2N6vXnHxvEJq1h2Orr6Sj/Qih5uey3lLRvb9ypHplJpd6lZSWseWcC0Zuz7Vzsbi7Bm8kf6ovOR26qMkcYOjiJNXD2k7JplJZvzlyWXc6MQ1PYzfdRKpyCaa7MevtjKdednYt9rQopv4JzJEd02u+esc+zO57sGs/SyqVwuQQ9lA87698Ub0yk029xnp+PnaeXQgsAHbE4m4L8CAw23+8ETgInJn2/LPwRvWaDpEj09cKra9hV3xwQoPXzliKXfMpTNOzBZ1CmQvTdRCz515vzv+S8wvdHZFJlY+gvxeoxQvzM/EOvnb6j/8E3AXcFou7S2NxdwGwFbh7tAOxMn6m6RmomA/zNk3I9m3FAuzaz3oXFWuZ3tf5N50HvLBfcoF36WeRIpHzrptoxOkFeod+jsXdo4CNRpwW/+evAzXADrwPlvvwplxKHpjBHmh+DrviEu/yA3m84Jktq8auuxra38I0POXNZpnmTGc97PsVdvUVMNhd0G8QE5kseb96ZTTiPM3b++GJRpxB4Av+TSbCoZdh3mbs8ou977nNA1s6E7vuGuhqwNQ/GoiQH2KOvQUHnsDWXg6DvV74iwSYJrgGgMF6AV/zbuys03Leng1XYNdeDQPtmLpfB+IrDIczba9imp/Hnv5pbNXiQndHZEIp6APC9LRA8/PY2stzurqlDVd4u2tSCcye+yf1AmWTrnkbHPkLdu1nsOVT+XwAkdwo6APEND0H/UextVFsFjtbvJC/BqzF7P7FtPrawmwYwBz4LXQewK69ekpe/lkkHxT0AWKwmH2/gsqF2FUfzWiHiy2rxm643vvqv92/wCSL45pzBoupi0GiE7vmSqzRfwkJHr2rA8YkOjG7fgFz12NXXDqukb2dtQrr3AC9rd6VKIsk5IcYm8TsfQDKZmf8ASkyHSjoA8j0tWJ23wPzHOy6q7GlM0/5PBuu5GjFZlKnfwZat2P2PYhJDU5yb6cGM9iD2f1LmLsBFr2n0N0RySt9OXhAme4m2Pkj7+DsGX+HPfIG5vg+GOiE8jnYmaug5gz6bC+ht/4DuoZfXbr4mL5W2PcQds2noe8IpiM/F0ETKTQFfYCZRBfs+jnMrsUuOAe7/BIomwUDx6GnhdC+h1iyoJLGnibtrvCZjr3Q+BR29SfgzZ9ielsL3SWRnCnoA84AHK/DHK8DvG+GGtprb0IhzILNBerZFHboj1BRgz39KnDv9s4+FpnGtI++yATpDNeJ4k27fBwGjmNXfyKrqaoiU4mCXuQUjE1h9j7ojeyXRwrdHZGcKOhFRmAGu71plwvPw87dUOjuiGRNQS8yCtPdiDn4W+xpH8dW1BS6OyJZUdCLjKV1O7S7pFZfSUrzF2QaUtCLjMEApv5xSPXTVnWWpqLKtKOgFxkHYwcJ7XuI/vA87OL3Fbo7IhlR0IuMkxnooKb3Fe+rCGevKXR3RMZNQS+SgcrBNkzTs9jVl2PLq8deQWQKUNCLZMgcehE667FrPoUN6eCsTH0KepEMGcDU/QZMWJc1lmlBQS+SBZMawOy9H6rXwYKzC90dkVEp6EWyZPqOYPY/jF3xQeyMZYXujsiIFPQiOTDtb8Lhl72vISyZUejuiJySgl4kR6bhKeg7gl1zha50KVOSgl4kRye+lL18nq50KVOSgl4kD3SlS5nKFPQieeJd6fJJ/0qX8wvdHZETFPQi+dT6Chx7yz+ZqqzQvREBFPQieeVd6fJRsEnsaR/TyVQyJSjoRfLMpAa9k6lm18Livyp0d0QU9CITwfQfw9T9Grv0/djZqwvdHSlyCnqRCWI69mCansGuvgJbPq/Q3ZEipqAXmUgtL0DHHuzaq7Dh8kL3RopUztdYjcXdcuB7wAeAhUAz8P1oxPk3v70EuB24Fu+D5X7g5mjE6c/1tUWmOgOw/xHshuuwq6+A3fdidIhWJlk+RvQlQAtwKTAb+DTwL7G4+xm//UvARcBmYC2wCfhGHl5XZFowdhCz5z6oWqQzZ6Ugch7RRyNON/C/0ha9Fou7jwDnA/cCNwG3RCNOE0As7m4F7onF3VujESeV6+uLTAcm0Ql77seu/xy25zDm6BuF7pIUkbx/PY6/q+YC4JuxuFsNrABeTXvKdmBoef1I2zGhECaU2R8cQ8/PdL1ipXplJtd6md5mUgcex552GWbgGKanKZ/dm3L0/spMLvUaa52J+B60O4AO4GfAIn9ZR1r7Mf9+1mgbWVbrkEwms+rAstqNWa1XrFSvzORWL0t7op6udZ9lUfc2SlM9eevXVKX3V2ayqVc4HB61Pa9BH4u7t+ON5i+ORpyBWNzt9JvmAG3+46FvVO4cvn66xjqXwcRARq9vQiGW1W6ksW4nNqW9QmNRvTKTr3pZ3sDWXk5T5VmEdv0MM9ibx15OHXp/ZSaXepWUlrHlnAtGbs+1c0Nicfff8GbeXByNOG0A0YhzLBZ3DwJnAnv9p56FN6o/ONr2bCqV9Zsjl3WLkeqVmbzUa18M1l1DavWnMLt+jkkN5qdzU5DeX5nJpl5jPT8vO89icfcO4BK8kG8d1nwXcFss7i6Nxd0FwFbgbh2IlWJmbNKbiROuwNZ+Ql9YIhMqH/PoVwF/D/QDdbG4O9T0XDTifAT4OlAD7MD7YLkPb8qlSFEzyT7YfQ92ww3YlZfCgScU9zIh8jG9sh5Gfn9GI84g8AX/JiJpzEAH7LkXu/5a6O+AQy8WuksSQJr3JFJgpqcFs/cB7LL3Y+efUejuSAAp6EWmAHN8H2b/w9hVl2Gr1xe6OxIwCnqRKcIcfQNz4Anvapdz1hS6OxIgCnqRKcS0vYppfAq75krsrNMK3R0JCAW9yBRjDv0R07TNu7Tx7NpCd0cCQEEvMgWZlj9gmp7Dnn6VduNIzhT0IlOUaXkB0/g0ds2ndIBWcqKgF5nCzKGXMAd/5x2grTmz0N2RaWoirl4pInlkWl+BwV5sbRRKqqDlDzqDVjKioBeZBkz7Tkj2Ytd8CspmwYHf6isJZdy060ZkmjDH6zBv/T+oXo9d+xl92biMm4JeZBoxPS0Y98dQUuVdDK18bqG7JNOAgl5kmjGJLsxbP4Pew1jnb7DV6wrdJZniFPQi05BJDWL2PYRpfg67+pOkll+MNfrvLKemg7Ei05QBOPRH6GrErvkkduYKqIth+tsL3TWZYjQEEJnmTHcjZuePINGF3XgTdsEWzceRd9CIXiQAzGAP7H0A5p+BXXEpVK+D+se8LzaRoqcRvUhAGMAc+Qtm551gU9jNf4tdcj7WhAvdNSkwjehFAsYMHIc9v/Tm26+8FOZthoanoGO3zqgtUgp6kQAyAMfeguP7vFH96iugpxkansJ0Nxa6ezLJFPQiAWZSCe8KmIdfwS69ELvhOuzxOkzz89B1UCP8IqGgFykCJtGJqX8U2/IidvF7sev+qzfCb3nR26VjU4XuokwgBb1IETH9RzH1j2CbnsUuOg972scglcC2/RnT9ppm6QSUgl6kCJlEJ6bh99imZ2DuRuyCs7BLL8B2HvSulNn+JibRVehuFg2LgZIZJEIzJmT7CnqRImZSg3DkdcyR17EVNdh5G7ELzoYVl2K7GzEde+H4Puhu1mWRM2QBQmVQOtO/zYDSmVj//h3LS6qwJsThZDfwUt77oqAXEQBMXxum6Vls07NQuRDmnI6dsxqWXACpAWxXI6a7AboavOBPDRS6y5PKG3VXeF/+UlLp33uP7fBlpVVQMhPCpd7KqQQkuiDR7d93Yboa3rHMJHtYuvI0JmJOlIJeRN7BAPQeht7DmJY/YENlMGslduZy7OxaWHw+hEux/cegx39efzsM3RJdU3Y2z4lRdkkFhCv8+0oIl0NJBTZc6S8r95afeJ4f5Mb/zVKDMNgDg73+vf+4pxmT8Jf5gU6iC1IDY9bEhEIT9leTgl5ERmVSA9CxB9OxB8C7Smb5PKhaiK1cCFWLsHPXQ/lcCJVAKon1A+5weYjUypXYZB8kBzDJfkgOQLIfUv3eSNdawJ763oS8WygMJv0W8l4rVAqhMmy4zHscLvOCPFTmPR5afuLncm/dIckBSPbBYJ93P/R4sBf62zFDj4eHeioxZT/MTkVBLyIZMTYFfW3Q14Zh54nlFqB0FpTN9r7usGwWpQtq6bNJKJkB5fO8QA6XQ6j87RDG+CNl/3745ZZTSbBpt/SfUwkvrE/cD3i7QlLt3gdUcuDt9pT/AZMW6sUyrVRBLyJ5YQASnd6t29sVMXd2P90H38CmMgtUe2KLdlqNnKcqBb2ITDleuGuWT77o6pUiIgGnoBcRCTgFvYhIwE3KPvpY3C0BbgeuxftwuR+4ORpx+ifj9UVEitlkjei/BFwEbAbWApuAb0zSa4uIFLXJmnVzE3BLNOI0AcTi7lbgnljcvTUacU4576q0rBwTyuxzyIRChMNhSssrMp7OVYxUr8yoXplRvTKTS71KSkpHb8+lY+MRi7vVwArg1bTF24Gh5fXDVpkN8JErbsj6Nbecc0HW6xYj1SszqldmVK/M5Fiv2cBJ15qejBH9LP8+/cWPDWtL1wCsBI5PZKdERAJmNl5+nmQygr7Tv58DtPmPq4e1nfBPn/+QBQ5OQr9ERIJkxG+NmfCDsdGIcwwvuM9MW3wW3qhegS4iMsEm62DsXcBtsbj7ApAAtgJ3j3QgVkRE8meygv7rQA2wA++viPvwplyKiMgEM9bqwkEiIkE27a5eGYu7lcBfgJpoxKn2l4165m2xnZkbi7s/Aa4B0r/rLRKNOC/77arXKcTi7seBLwPr8CYKfDsacb6ler1TLO4O/9bwcsCNRpx3+e2q1zCxuLsU+C7eiaMG2Ib3OzdMRr2m47VuvsLJU4jGOvO2GM/M/UE04sxMu72c1qZ6DROLux8Gfgj8I96ssPXAY36z6pVm2PtqJuAC96Q9RfU62Q+AUqAW7/yhbuBOv23C6zWtgj4Wd7cAH+XkX/Im4GvRiNMUjTiteAd7/zoWd0PjbC82qtfJvor3O/8+GnEGoxHneDTivOG3qV4jiMXd84CNwE/SFqteJ1sN3BuNOJ3RiNMD/Bw4w2+b8HpNm103/p8vdwI3D1s+6pm3sbjbMVo7J5+ZGxTXxeLudUAz8GPgO9GIk1K9ThaLuzOAs/Euy7ETmA/8Afgi3ol7qtfIbgQeS7u8id5fp/Zt4KpY3H0ESOLthnlksuo1nT5B/wF4PRpxnh62fKwzbzM9MzcI7sDb9bAAbzTwRf8GqtepzMXbb3o93l+MtcBh4AFUrxHF4m4V8Fm86dNDVK9Tex6YBxzF+33XAf/MJNVrWgR9LO6uwRvJ33qK5vQzb4ekn3k7VnvgRCPO9mjEaY1GnGQ04rwA/CvwGb9Z9TrZ0O91RzTi7Pf/tL4NOAdv9AWq16lcBfQAj6Qt0/trGH8Xy5PAC8BMvID+Hd4xoEmp17QIeuBCvNHpjljcbQEeBGb7jzcyypm3OjMXgBMnpo1Vj2KsVzTidOD9CZw+1zj9sep1ajcBP41GnMGhBXp/ndI8YBXeQKLHH0h8F3gP3u7zCa/XtJhH70+pTP9Eex9wN97uiaN4fwJ9AvgY3pm3vwZejEacW/z1//do7UETi7tXAY/jfeKfjTcd6/vRiPMtv33UehRbvQBicfef8aakXga04u1TPTcacc5VvU4Wi7vr8WbbbIhGnF3D2lSvYWJxdzfe/8Mv+4u+AlwdjTgrJqNe0+JgbDTi9AK9Qz/H4u5RwEYjTov/81hn3hbbmbn/Hfi/eP++jXhTu25Pa1e9TvZNvH312/F+523AlX6b6nWyG4Hnhoe8T/U62eXOAmWUAAAATklEQVTAd/D+PxrgNSDqt014vabFiF5ERLI3XfbRi4hIlhT0IiIBp6AXEQk4Bb2ISMAp6EVEAk5BLyIScAp6EZGAU9CLiAScgl5EJOD+P1OHoKM8Ua/NAAAAAElFTkSuQmCC\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "dark" }, "output_type": "display_data" } ], "source": [ "lam_inv = np.linspace(1/400.0, 1/800.0, num=num_lpoints, endpoint=True)\n", "lams = 1.0/lam_inv\n", "\n", "plt.plot( lams, y_train[ np.random.randint(0, 9000) ] )\n", "plt.ylim( [0, 100])" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "ExecuteTime": { "end_time": "2018-12-24T07:08:16.617970Z", "start_time": "2018-12-24T07:08:16.058432Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n", "Layer (type) Output Shape Param # \n", "=================================================================\n", "first (Dense) (None, 512) 8704 \n", "_________________________________________________________________\n", "activation_17 (Activation) (None, 512) 0 \n", "_________________________________________________________________\n", "H0 (Dense) (None, 512) 262656 \n", "_________________________________________________________________\n", "activation_18 (Activation) (None, 512) 0 \n", "_________________________________________________________________\n", "H1 (Dense) (None, 512) 262656 \n", "_________________________________________________________________\n", "activation_19 (Activation) (None, 512) 0 \n", "_________________________________________________________________\n", "H2 (Dense) (None, 512) 262656 \n", "_________________________________________________________________\n", "activation_20 (Activation) (None, 512) 0 \n", "_________________________________________________________________\n", "last (Dense) (None, 128) 65664 \n", "=================================================================\n", "Total params: 862,336\n", "Trainable params: 862,336\n", "Non-trainable params: 0\n", "_________________________________________________________________\n" ] } ], "source": [ "import scnets as scn\n", "model = scn.fullycon(in_size=16, \n", " out_size=128, \n", " batch_size=64,\n", " N_hidden=3, \n", " N_neurons=512, \n", " N_gpus=3)\n", "\n", "model.summary() \n" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "ExecuteTime": { "end_time": "2018-12-24T08:41:46.256500Z", "start_time": "2018-12-24T07:55:13.641148Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train on 180000 samples, validate on 20000 samples\n", "Epoch 1/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 7099.9878 - calc_mre_K: 2.6003 - val_loss: 8302.0331 - val_calc_mre_K: 3.0422\n", "Epoch 2/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 7110.3147 - calc_mre_K: 2.6040 - val_loss: 8196.3713 - val_calc_mre_K: 3.0037\n", "Epoch 3/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7088.3919 - calc_mre_K: 2.5961 - val_loss: 7958.5186 - val_calc_mre_K: 2.9162\n", "Epoch 4/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7090.3141 - calc_mre_K: 2.5968 - val_loss: 7992.5676 - val_calc_mre_K: 2.9288\n", "Epoch 5/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7084.5473 - calc_mre_K: 2.5947 - val_loss: 7959.9942 - val_calc_mre_K: 2.9168\n", "Epoch 6/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7087.6522 - calc_mre_K: 2.5958 - val_loss: 7904.4409 - val_calc_mre_K: 2.8966\n", "Epoch 7/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7070.5508 - calc_mre_K: 2.5895 - val_loss: 8158.7949 - val_calc_mre_K: 2.9896\n", "Epoch 8/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 7078.3557 - calc_mre_K: 2.5923 - val_loss: 7896.0569 - val_calc_mre_K: 2.8935\n", "Epoch 9/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7067.8534 - calc_mre_K: 2.5885 - val_loss: 7974.0039 - val_calc_mre_K: 2.9222\n", "Epoch 10/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 7060.3409 - calc_mre_K: 2.5857 - val_loss: 7832.0003 - val_calc_mre_K: 2.8701\n", "Epoch 11/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 7052.0339 - calc_mre_K: 2.5829 - val_loss: 7875.6581 - val_calc_mre_K: 2.8859\n", "Epoch 12/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7065.0015 - calc_mre_K: 2.5876 - val_loss: 8002.7019 - val_calc_mre_K: 2.9325\n", "Epoch 13/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 7040.1258 - calc_mre_K: 2.5784 - val_loss: 8227.0031 - val_calc_mre_K: 3.0147\n", "Epoch 14/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 7051.4860 - calc_mre_K: 2.5826 - val_loss: 8093.0015 - val_calc_mre_K: 2.9656\n", "Epoch 15/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 7028.3113 - calc_mre_K: 2.5741 - val_loss: 8059.1222 - val_calc_mre_K: 2.9533\n", "Epoch 16/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7034.2063 - calc_mre_K: 2.5762 - val_loss: 7952.5174 - val_calc_mre_K: 2.9143\n", "Epoch 17/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 7018.3979 - calc_mre_K: 2.5705 - val_loss: 7999.9423 - val_calc_mre_K: 2.9318\n", "Epoch 18/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 7009.9372 - calc_mre_K: 2.5673 - val_loss: 8009.2476 - val_calc_mre_K: 2.9350\n", "Epoch 19/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 7000.1075 - calc_mre_K: 2.5638 - val_loss: 8276.7471 - val_calc_mre_K: 3.0330\n", "Epoch 20/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 7013.6226 - calc_mre_K: 2.5687 - val_loss: 7776.5133 - val_calc_mre_K: 2.8497\n", "Epoch 21/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6988.1233 - calc_mre_K: 2.5594 - val_loss: 7932.4221 - val_calc_mre_K: 2.9069\n", "Epoch 22/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 7001.3027 - calc_mre_K: 2.5642 - val_loss: 8089.4508 - val_calc_mre_K: 2.9644\n", "Epoch 23/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6982.6276 - calc_mre_K: 2.5574 - val_loss: 8010.3768 - val_calc_mre_K: 2.9354\n", "Epoch 24/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6981.2012 - calc_mre_K: 2.5568 - val_loss: 8043.1651 - val_calc_mre_K: 2.9473\n", "Epoch 25/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6975.5134 - calc_mre_K: 2.5547 - val_loss: 7673.5979 - val_calc_mre_K: 2.8120\n", "Epoch 26/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6965.4576 - calc_mre_K: 2.5511 - val_loss: 7772.9833 - val_calc_mre_K: 2.8483\n", "Epoch 27/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6961.2040 - calc_mre_K: 2.5496 - val_loss: 7974.7777 - val_calc_mre_K: 2.9226\n", "Epoch 28/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6974.4708 - calc_mre_K: 2.5544 - val_loss: 8011.7313 - val_calc_mre_K: 2.9359\n", "Epoch 29/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6959.6734 - calc_mre_K: 2.5490 - val_loss: 7824.1970 - val_calc_mre_K: 2.8672\n", "Epoch 30/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6964.1885 - calc_mre_K: 2.5505 - val_loss: 7853.0341 - val_calc_mre_K: 2.8777\n", "Epoch 31/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6934.7727 - calc_mre_K: 2.5398 - val_loss: 7735.1516 - val_calc_mre_K: 2.8344\n", "Epoch 32/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6958.9001 - calc_mre_K: 2.5487 - val_loss: 7850.4633 - val_calc_mre_K: 2.8768\n", "Epoch 33/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6934.9780 - calc_mre_K: 2.5399 - val_loss: 7853.0112 - val_calc_mre_K: 2.8779\n", "Epoch 34/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6928.6110 - calc_mre_K: 2.5376 - val_loss: 7766.1565 - val_calc_mre_K: 2.8461\n", "Epoch 35/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6937.1697 - calc_mre_K: 2.5407 - val_loss: 7889.2375 - val_calc_mre_K: 2.8909\n", "Epoch 36/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6922.1167 - calc_mre_K: 2.5352 - val_loss: 8294.7321 - val_calc_mre_K: 3.0396\n", "Epoch 37/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6918.5657 - calc_mre_K: 2.5338 - val_loss: 7925.6814 - val_calc_mre_K: 2.9044\n", "Epoch 38/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6910.4561 - calc_mre_K: 2.5310 - val_loss: 7654.9729 - val_calc_mre_K: 2.8053\n", "Epoch 39/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6908.1185 - calc_mre_K: 2.5301 - val_loss: 7601.9860 - val_calc_mre_K: 2.7859\n", "Epoch 40/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6902.8513 - calc_mre_K: 2.5281 - val_loss: 7720.6251 - val_calc_mre_K: 2.8290\n", "Epoch 41/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6914.2222 - calc_mre_K: 2.5323 - val_loss: 8090.7705 - val_calc_mre_K: 2.9648\n", "Epoch 42/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6900.2506 - calc_mre_K: 2.5272 - val_loss: 7850.8699 - val_calc_mre_K: 2.8769\n", "Epoch 43/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6892.6605 - calc_mre_K: 2.5243 - val_loss: 7711.9717 - val_calc_mre_K: 2.8260\n", "Epoch 44/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6888.5774 - calc_mre_K: 2.5229 - val_loss: 7807.7505 - val_calc_mre_K: 2.8611\n", "Epoch 45/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6891.7760 - calc_mre_K: 2.5241 - val_loss: 8174.6846 - val_calc_mre_K: 2.9957\n", "Epoch 46/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6878.3412 - calc_mre_K: 2.5191 - val_loss: 8043.4201 - val_calc_mre_K: 2.9476\n", "Epoch 47/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6877.8314 - calc_mre_K: 2.5189 - val_loss: 7830.0886 - val_calc_mre_K: 2.8693\n", "Epoch 48/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6863.0152 - calc_mre_K: 2.5135 - val_loss: 7948.4689 - val_calc_mre_K: 2.9126\n", "Epoch 49/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6874.4221 - calc_mre_K: 2.5177 - val_loss: 7862.2067 - val_calc_mre_K: 2.8812\n", "Epoch 50/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6863.7506 - calc_mre_K: 2.5138 - val_loss: 7760.4736 - val_calc_mre_K: 2.8439\n", "Epoch 51/200\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "180000/180000 [==============================] - 14s 76us/step - loss: 6867.5228 - calc_mre_K: 2.5152 - val_loss: 7759.6516 - val_calc_mre_K: 2.8435\n", "Epoch 52/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6862.0412 - calc_mre_K: 2.5132 - val_loss: 7847.5048 - val_calc_mre_K: 2.8758\n", "Epoch 53/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6851.9832 - calc_mre_K: 2.5095 - val_loss: 7702.3974 - val_calc_mre_K: 2.8227\n", "Epoch 54/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6848.3869 - calc_mre_K: 2.5082 - val_loss: 7714.3371 - val_calc_mre_K: 2.8268\n", "Epoch 55/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6850.9434 - calc_mre_K: 2.5092 - val_loss: 7721.9557 - val_calc_mre_K: 2.8297\n", "Epoch 56/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6841.0665 - calc_mre_K: 2.5054 - val_loss: 7861.2203 - val_calc_mre_K: 2.8808\n", "Epoch 57/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6845.7766 - calc_mre_K: 2.5072 - val_loss: 7717.1635 - val_calc_mre_K: 2.8279\n", "Epoch 58/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6835.7648 - calc_mre_K: 2.5036 - val_loss: 7663.9810 - val_calc_mre_K: 2.8085\n", "Epoch 59/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6822.6891 - calc_mre_K: 2.4987 - val_loss: 7940.1643 - val_calc_mre_K: 2.9096\n", "Epoch 60/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6820.7750 - calc_mre_K: 2.4981 - val_loss: 7828.2685 - val_calc_mre_K: 2.8688\n", "Epoch 61/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6834.9354 - calc_mre_K: 2.5033 - val_loss: 7824.3697 - val_calc_mre_K: 2.8674\n", "Epoch 62/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6825.2897 - calc_mre_K: 2.4998 - val_loss: 7922.3464 - val_calc_mre_K: 2.9034\n", "Epoch 63/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6812.9849 - calc_mre_K: 2.4953 - val_loss: 7959.9627 - val_calc_mre_K: 2.9170\n", "Epoch 64/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6827.6530 - calc_mre_K: 2.5006 - val_loss: 7607.3291 - val_calc_mre_K: 2.7878\n", "Epoch 65/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6797.2376 - calc_mre_K: 2.4895 - val_loss: 7768.9983 - val_calc_mre_K: 2.8468\n", "Epoch 66/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6803.6693 - calc_mre_K: 2.4918 - val_loss: 7693.5868 - val_calc_mre_K: 2.8193\n", "Epoch 67/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6799.4362 - calc_mre_K: 2.4903 - val_loss: 7813.4609 - val_calc_mre_K: 2.8631\n", "Epoch 68/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6793.8019 - calc_mre_K: 2.4882 - val_loss: 7554.5192 - val_calc_mre_K: 2.7682\n", "Epoch 69/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6781.8687 - calc_mre_K: 2.4838 - val_loss: 7599.6239 - val_calc_mre_K: 2.7848\n", "Epoch 70/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6790.3035 - calc_mre_K: 2.4869 - val_loss: 7621.5495 - val_calc_mre_K: 2.7929\n", "Epoch 71/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6778.7803 - calc_mre_K: 2.4826 - val_loss: 8593.1654 - val_calc_mre_K: 3.1491\n", "Epoch 72/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6789.7105 - calc_mre_K: 2.4867 - val_loss: 7802.8288 - val_calc_mre_K: 2.8593\n", "Epoch 73/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6771.8280 - calc_mre_K: 2.4802 - val_loss: 7888.7872 - val_calc_mre_K: 2.8907\n", "Epoch 74/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6776.7983 - calc_mre_K: 2.4820 - val_loss: 7710.8632 - val_calc_mre_K: 2.8257\n", "Epoch 75/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6754.4425 - calc_mre_K: 2.4738 - val_loss: 7677.3155 - val_calc_mre_K: 2.8133\n", "Epoch 76/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6760.5672 - calc_mre_K: 2.4760 - val_loss: 8002.1879 - val_calc_mre_K: 2.9324\n", "Epoch 77/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6765.5668 - calc_mre_K: 2.4779 - val_loss: 7861.0321 - val_calc_mre_K: 2.8807\n", "Epoch 78/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6746.4606 - calc_mre_K: 2.4709 - val_loss: 7699.0299 - val_calc_mre_K: 2.8212\n", "Epoch 79/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6765.8151 - calc_mre_K: 2.4780 - val_loss: 8005.5933 - val_calc_mre_K: 2.9337\n", "Epoch 80/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6742.2366 - calc_mre_K: 2.4693 - val_loss: 7677.1655 - val_calc_mre_K: 2.8134\n", "Epoch 81/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6752.1178 - calc_mre_K: 2.4729 - val_loss: 8029.6324 - val_calc_mre_K: 2.9424\n", "Epoch 82/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6749.9855 - calc_mre_K: 2.4721 - val_loss: 7631.2121 - val_calc_mre_K: 2.7964\n", "Epoch 83/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6734.8218 - calc_mre_K: 2.4665 - val_loss: 7779.7796 - val_calc_mre_K: 2.8508\n", "Epoch 84/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6728.9338 - calc_mre_K: 2.4645 - val_loss: 7772.6415 - val_calc_mre_K: 2.8483\n", "Epoch 85/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6740.7611 - calc_mre_K: 2.4688 - val_loss: 7723.1116 - val_calc_mre_K: 2.8301\n", "Epoch 86/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6733.1725 - calc_mre_K: 2.4661 - val_loss: 7566.5807 - val_calc_mre_K: 2.7729\n", "Epoch 87/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6745.9367 - calc_mre_K: 2.4706 - val_loss: 7703.0090 - val_calc_mre_K: 2.8227\n", "Epoch 88/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6712.8064 - calc_mre_K: 2.4586 - val_loss: 7935.4137 - val_calc_mre_K: 2.9079\n", "Epoch 89/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6729.2436 - calc_mre_K: 2.4645 - val_loss: 7529.3015 - val_calc_mre_K: 2.7592\n", "Epoch 90/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6715.8521 - calc_mre_K: 2.4597 - val_loss: 7690.4235 - val_calc_mre_K: 2.8183\n", "Epoch 91/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6723.4227 - calc_mre_K: 2.4623 - val_loss: 7614.4126 - val_calc_mre_K: 2.7903\n", "Epoch 92/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6710.6667 - calc_mre_K: 2.4577 - val_loss: 7680.5015 - val_calc_mre_K: 2.8146\n", "Epoch 93/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6710.5442 - calc_mre_K: 2.4576 - val_loss: 7803.6383 - val_calc_mre_K: 2.8597\n", "Epoch 94/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6696.8179 - calc_mre_K: 2.4527 - val_loss: 7569.0176 - val_calc_mre_K: 2.7738\n", "Epoch 95/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6701.9745 - calc_mre_K: 2.4545 - val_loss: 7545.8799 - val_calc_mre_K: 2.7652\n", "Epoch 96/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6693.1059 - calc_mre_K: 2.4513 - val_loss: 7682.2530 - val_calc_mre_K: 2.8151\n", "Epoch 97/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6688.6574 - calc_mre_K: 2.4498 - val_loss: 7558.3737 - val_calc_mre_K: 2.7697\n", "Epoch 98/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6695.0895 - calc_mre_K: 2.4520 - val_loss: 7513.9831 - val_calc_mre_K: 2.7533\n", "Epoch 99/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6698.0173 - calc_mre_K: 2.4531 - val_loss: 7819.3289 - val_calc_mre_K: 2.8653\n", "Epoch 100/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6689.3852 - calc_mre_K: 2.4499 - val_loss: 7554.0840 - val_calc_mre_K: 2.7681\n", "Epoch 101/200\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "180000/180000 [==============================] - 14s 76us/step - loss: 6679.5679 - calc_mre_K: 2.4464 - val_loss: 7737.1785 - val_calc_mre_K: 2.8352\n", "Epoch 102/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6670.6785 - calc_mre_K: 2.4432 - val_loss: 7599.0685 - val_calc_mre_K: 2.7846\n", "Epoch 103/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6680.9516 - calc_mre_K: 2.4468 - val_loss: 7401.7399 - val_calc_mre_K: 2.7123\n", "Epoch 104/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6671.7837 - calc_mre_K: 2.4435 - val_loss: 7578.3705 - val_calc_mre_K: 2.7771\n", "Epoch 105/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6665.8595 - calc_mre_K: 2.4414 - val_loss: 7792.2259 - val_calc_mre_K: 2.8553\n", "Epoch 106/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6682.9109 - calc_mre_K: 2.4476 - val_loss: 7749.7058 - val_calc_mre_K: 2.8396\n", "Epoch 107/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6659.8448 - calc_mre_K: 2.4392 - val_loss: 7922.2356 - val_calc_mre_K: 2.9031\n", "Epoch 108/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6659.9848 - calc_mre_K: 2.4393 - val_loss: 7713.1598 - val_calc_mre_K: 2.8263\n", "Epoch 109/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6666.3733 - calc_mre_K: 2.4416 - val_loss: 7607.7003 - val_calc_mre_K: 2.7877\n", "Epoch 110/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6657.9852 - calc_mre_K: 2.4385 - val_loss: 7747.7073 - val_calc_mre_K: 2.8390\n", "Epoch 111/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6655.9008 - calc_mre_K: 2.4377 - val_loss: 7562.5887 - val_calc_mre_K: 2.7713\n", "Epoch 112/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6640.3608 - calc_mre_K: 2.4320 - val_loss: 7585.6801 - val_calc_mre_K: 2.7796\n", "Epoch 113/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6656.7902 - calc_mre_K: 2.4380 - val_loss: 7713.1042 - val_calc_mre_K: 2.8266\n", "Epoch 114/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6642.5611 - calc_mre_K: 2.4328 - val_loss: 7482.7650 - val_calc_mre_K: 2.7420\n", "Epoch 115/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6651.6710 - calc_mre_K: 2.4361 - val_loss: 7812.7017 - val_calc_mre_K: 2.8630\n", "Epoch 116/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6621.9042 - calc_mre_K: 2.4252 - val_loss: 7522.0455 - val_calc_mre_K: 2.7564\n", "Epoch 117/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6621.0187 - calc_mre_K: 2.4250 - val_loss: 7861.0635 - val_calc_mre_K: 2.8808\n", "Epoch 118/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6631.8639 - calc_mre_K: 2.4289 - val_loss: 7707.9886 - val_calc_mre_K: 2.8247\n", "Epoch 119/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6616.0955 - calc_mre_K: 2.4231 - val_loss: 7521.2598 - val_calc_mre_K: 2.7562\n", "Epoch 120/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6622.3670 - calc_mre_K: 2.4254 - val_loss: 7513.6696 - val_calc_mre_K: 2.7532\n", "Epoch 121/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6619.4877 - calc_mre_K: 2.4243 - val_loss: 7543.0488 - val_calc_mre_K: 2.7641\n", "Epoch 122/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6618.4689 - calc_mre_K: 2.4239 - val_loss: 7570.1745 - val_calc_mre_K: 2.7740\n", "Epoch 123/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6614.6816 - calc_mre_K: 2.4226 - val_loss: 7522.8252 - val_calc_mre_K: 2.7565\n", "Epoch 124/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6620.1709 - calc_mre_K: 2.4246 - val_loss: 7729.4280 - val_calc_mre_K: 2.8324\n", "Epoch 125/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6619.3056 - calc_mre_K: 2.4242 - val_loss: 7510.7295 - val_calc_mre_K: 2.7522\n", "Epoch 126/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6613.3750 - calc_mre_K: 2.4221 - val_loss: 7526.6338 - val_calc_mre_K: 2.7579\n", "Epoch 127/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6601.8710 - calc_mre_K: 2.4179 - val_loss: 7723.0214 - val_calc_mre_K: 2.8301\n", "Epoch 128/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6605.8103 - calc_mre_K: 2.4193 - val_loss: 7538.5741 - val_calc_mre_K: 2.7624\n", "Epoch 129/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6602.8332 - calc_mre_K: 2.4183 - val_loss: 7865.8780 - val_calc_mre_K: 2.8823\n", "Epoch 130/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6602.2529 - calc_mre_K: 2.4181 - val_loss: 7641.1493 - val_calc_mre_K: 2.8000\n", "Epoch 131/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6594.2067 - calc_mre_K: 2.4152 - val_loss: 7393.9496 - val_calc_mre_K: 2.7094\n", "Epoch 132/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6597.4194 - calc_mre_K: 2.4162 - val_loss: 7549.6489 - val_calc_mre_K: 2.7663\n", "Epoch 133/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6589.3893 - calc_mre_K: 2.4133 - val_loss: 7551.1380 - val_calc_mre_K: 2.7671\n", "Epoch 134/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6585.0070 - calc_mre_K: 2.4118 - val_loss: 7680.2262 - val_calc_mre_K: 2.8144\n", "Epoch 135/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6580.5867 - calc_mre_K: 2.4101 - val_loss: 7662.6057 - val_calc_mre_K: 2.8079\n", "Epoch 136/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6584.5403 - calc_mre_K: 2.4116 - val_loss: 7616.7852 - val_calc_mre_K: 2.7912\n", "Epoch 137/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6576.9014 - calc_mre_K: 2.4088 - val_loss: 7602.7721 - val_calc_mre_K: 2.7860\n", "Epoch 138/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6586.3951 - calc_mre_K: 2.4122 - val_loss: 7406.2313 - val_calc_mre_K: 2.7139\n", "Epoch 139/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6580.6517 - calc_mre_K: 2.4101 - val_loss: 7446.2008 - val_calc_mre_K: 2.7286\n", "Epoch 140/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6571.3992 - calc_mre_K: 2.4067 - val_loss: 7661.9602 - val_calc_mre_K: 2.8076\n", "Epoch 141/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6571.0040 - calc_mre_K: 2.4066 - val_loss: 7618.4710 - val_calc_mre_K: 2.7916\n", "Epoch 142/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6575.7832 - calc_mre_K: 2.4083 - val_loss: 7486.0405 - val_calc_mre_K: 2.7432\n", "Epoch 143/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6560.1383 - calc_mre_K: 2.4027 - val_loss: 7526.6525 - val_calc_mre_K: 2.7579\n", "Epoch 144/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6551.2769 - calc_mre_K: 2.3993 - val_loss: 7583.4914 - val_calc_mre_K: 2.7791\n", "Epoch 145/200\n", "180000/180000 [==============================] - 14s 75us/step - loss: 6562.3775 - calc_mre_K: 2.4035 - val_loss: 7434.2014 - val_calc_mre_K: 2.7242\n", "Epoch 146/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6558.2140 - calc_mre_K: 2.4020 - val_loss: 7679.9207 - val_calc_mre_K: 2.8142\n", "Epoch 147/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6550.4137 - calc_mre_K: 2.3991 - val_loss: 7443.1142 - val_calc_mre_K: 2.7276\n", "Epoch 148/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6561.1909 - calc_mre_K: 2.4030 - val_loss: 7410.3766 - val_calc_mre_K: 2.7156\n", "Epoch 149/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6545.9555 - calc_mre_K: 2.3974 - val_loss: 7368.9968 - val_calc_mre_K: 2.7002\n", "Epoch 150/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6563.2443 - calc_mre_K: 2.4037 - val_loss: 7804.3743 - val_calc_mre_K: 2.8599\n", "Epoch 151/200\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "180000/180000 [==============================] - 14s 77us/step - loss: 6547.5754 - calc_mre_K: 2.3980 - val_loss: 7522.0964 - val_calc_mre_K: 2.7564\n", "Epoch 152/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6541.8411 - calc_mre_K: 2.3959 - val_loss: 7857.3824 - val_calc_mre_K: 2.8792\n", "Epoch 153/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6542.3811 - calc_mre_K: 2.3961 - val_loss: 7637.7159 - val_calc_mre_K: 2.7986\n", "Epoch 154/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6546.9450 - calc_mre_K: 2.3978 - val_loss: 7551.1645 - val_calc_mre_K: 2.7671\n", "Epoch 155/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6535.2760 - calc_mre_K: 2.3935 - val_loss: 7551.6044 - val_calc_mre_K: 2.7673\n", "Epoch 156/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6534.2796 - calc_mre_K: 2.3931 - val_loss: 7306.6364 - val_calc_mre_K: 2.6774\n", "Epoch 157/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6523.6742 - calc_mre_K: 2.3893 - val_loss: 7547.2222 - val_calc_mre_K: 2.7654\n", "Epoch 158/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6534.6232 - calc_mre_K: 2.3933 - val_loss: 7485.2916 - val_calc_mre_K: 2.7429\n", "Epoch 159/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6522.3733 - calc_mre_K: 2.3888 - val_loss: 7551.3576 - val_calc_mre_K: 2.7671\n", "Epoch 160/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6535.4804 - calc_mre_K: 2.3936 - val_loss: 7375.9345 - val_calc_mre_K: 2.7027\n", "Epoch 161/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6523.8203 - calc_mre_K: 2.3893 - val_loss: 7492.0995 - val_calc_mre_K: 2.7455\n", "Epoch 162/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6531.0059 - calc_mre_K: 2.3920 - val_loss: 7414.1955 - val_calc_mre_K: 2.7169\n", "Epoch 163/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6520.9405 - calc_mre_K: 2.3882 - val_loss: 7486.3791 - val_calc_mre_K: 2.7433\n", "Epoch 164/200\n", "180000/180000 [==============================] - 14s 75us/step - loss: 6523.7890 - calc_mre_K: 2.3892 - val_loss: 7436.5905 - val_calc_mre_K: 2.7252\n", "Epoch 165/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6512.7467 - calc_mre_K: 2.3853 - val_loss: 7421.0140 - val_calc_mre_K: 2.7192\n", "Epoch 166/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6522.1952 - calc_mre_K: 2.3886 - val_loss: 7452.3864 - val_calc_mre_K: 2.7310\n", "Epoch 167/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6510.2451 - calc_mre_K: 2.3843 - val_loss: 7549.6243 - val_calc_mre_K: 2.7664\n", "Epoch 168/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6505.9163 - calc_mre_K: 2.3828 - val_loss: 7628.2356 - val_calc_mre_K: 2.7954\n", "Epoch 169/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6516.7098 - calc_mre_K: 2.3867 - val_loss: 7515.7280 - val_calc_mre_K: 2.7539\n", "Epoch 170/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6503.6826 - calc_mre_K: 2.3819 - val_loss: 7377.4004 - val_calc_mre_K: 2.7032\n", "Epoch 171/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6487.3929 - calc_mre_K: 2.3760 - val_loss: 7340.5011 - val_calc_mre_K: 2.6897\n", "Epoch 172/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6492.9036 - calc_mre_K: 2.3780 - val_loss: 8147.3081 - val_calc_mre_K: 2.9854\n", "Epoch 173/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6500.9393 - calc_mre_K: 2.3810 - val_loss: 7529.5955 - val_calc_mre_K: 2.7591\n", "Epoch 174/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6490.2481 - calc_mre_K: 2.3770 - val_loss: 7500.4239 - val_calc_mre_K: 2.7484\n", "Epoch 175/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6492.4610 - calc_mre_K: 2.3779 - val_loss: 7429.0373 - val_calc_mre_K: 2.7222\n", "Epoch 176/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6491.3703 - calc_mre_K: 2.3775 - val_loss: 7849.9025 - val_calc_mre_K: 2.8768\n", "Epoch 177/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6493.4181 - calc_mre_K: 2.3782 - val_loss: 7571.6186 - val_calc_mre_K: 2.7745\n", "Epoch 178/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6495.6656 - calc_mre_K: 2.3790 - val_loss: 7518.4636 - val_calc_mre_K: 2.7549\n", "Epoch 179/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6480.1927 - calc_mre_K: 2.3732 - val_loss: 7468.6425 - val_calc_mre_K: 2.7367\n", "Epoch 180/200\n", "180000/180000 [==============================] - 14s 80us/step - loss: 6491.7191 - calc_mre_K: 2.3775 - val_loss: 7562.2013 - val_calc_mre_K: 2.7711\n", "Epoch 181/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6482.9569 - calc_mre_K: 2.3743 - val_loss: 7480.9571 - val_calc_mre_K: 2.7412\n", "Epoch 182/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6473.0910 - calc_mre_K: 2.3708 - val_loss: 7710.0109 - val_calc_mre_K: 2.8253\n", "Epoch 183/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6481.8012 - calc_mre_K: 2.3739 - val_loss: 7641.5195 - val_calc_mre_K: 2.8000\n", "Epoch 184/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6474.4546 - calc_mre_K: 2.3712 - val_loss: 7716.8835 - val_calc_mre_K: 2.8277\n", "Epoch 185/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6476.7034 - calc_mre_K: 2.3721 - val_loss: 7534.1252 - val_calc_mre_K: 2.7608\n", "Epoch 186/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6470.5269 - calc_mre_K: 2.3697 - val_loss: 7475.4594 - val_calc_mre_K: 2.7392\n", "Epoch 187/200\n", "180000/180000 [==============================] - 14s 78us/step - loss: 6464.0561 - calc_mre_K: 2.3674 - val_loss: 7477.4329 - val_calc_mre_K: 2.7398\n", "Epoch 188/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6463.6265 - calc_mre_K: 2.3671 - val_loss: 7562.6554 - val_calc_mre_K: 2.7711\n", "Epoch 189/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6463.0398 - calc_mre_K: 2.3671 - val_loss: 7368.0955 - val_calc_mre_K: 2.6999\n", "Epoch 190/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6447.9319 - calc_mre_K: 2.3615 - val_loss: 7450.0313 - val_calc_mre_K: 2.7299\n", "Epoch 191/200\n", "180000/180000 [==============================] - 14s 75us/step - loss: 6464.8206 - calc_mre_K: 2.3677 - val_loss: 7387.9698 - val_calc_mre_K: 2.7073\n", "Epoch 192/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6459.9742 - calc_mre_K: 2.3660 - val_loss: 7294.1589 - val_calc_mre_K: 2.6729\n", "Epoch 193/200\n", "180000/180000 [==============================] - 14s 75us/step - loss: 6467.4806 - calc_mre_K: 2.3686 - val_loss: 7779.8633 - val_calc_mre_K: 2.8506\n", "Epoch 194/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6449.6030 - calc_mre_K: 2.3621 - val_loss: 7397.3320 - val_calc_mre_K: 2.7106\n", "Epoch 195/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6452.4974 - calc_mre_K: 2.3632 - val_loss: 7490.6713 - val_calc_mre_K: 2.7449\n", "Epoch 196/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6450.5287 - calc_mre_K: 2.3625 - val_loss: 7567.7744 - val_calc_mre_K: 2.7731\n", "Epoch 197/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6449.2848 - calc_mre_K: 2.3621 - val_loss: 7688.1467 - val_calc_mre_K: 2.8171\n", "Epoch 198/200\n", "180000/180000 [==============================] - 14s 76us/step - loss: 6453.6967 - calc_mre_K: 2.3637 - val_loss: 7366.5965 - val_calc_mre_K: 2.6992\n", "Epoch 199/200\n", "180000/180000 [==============================] - 14s 79us/step - loss: 6446.5738 - calc_mre_K: 2.3610 - val_loss: 7368.0158 - val_calc_mre_K: 2.7000\n", "Epoch 200/200\n", "180000/180000 [==============================] - 14s 77us/step - loss: 6436.0559 - calc_mre_K: 2.3571 - val_loss: 7482.3732 - val_calc_mre_K: 2.7418\n" ] } ], "source": [ "history = model.fit(x_train, y_train,\n", " batch_size=64,\n", " epochs=200, \n", " verbose=1,\n", " validation_data=(x_test, y_test))\n" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "ExecuteTime": { "end_time": "2018-12-24T08:43:50.327911Z", "start_time": "2018-12-24T08:43:50.304722Z" } }, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 50, "metadata": { "ExecuteTime": { "end_time": "2018-12-24T08:48:58.045055Z", "start_time": "2018-12-24T08:48:57.919927Z" } }, "outputs": [ { "data": { "text/plain": [ "(0, 100)" ] }, "execution_count": 50, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "dark" }, "output_type": "display_data" } ], "source": [ "numbr = np.random.randint(0,2000)\n", "yz = model.predict(x_test[numbr:numbr+3])\n", "plt.plot(yz[0])\n", "plt.plot(y_test[numbr])\n", "plt.ylim([0,100])" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "hide_input": false, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 2 }