{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-11T05:53:10.003222Z",
     "start_time": "2018-09-11T05:53:09.983926Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/hegder/anaconda3/envs/deep/lib/python3.6/site-packages/ipyparallel/client/client.py:459: RuntimeWarning: \n",
      "            Controller appears to be listening on localhost, but not on this machine.\n",
      "            If this is true, you should specify Client(...,sshserver='you@Gamma')\n",
      "            or instruct your controller to listen on an external IP.\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0, 1, 2, 3, 4, 5, 6, 7]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import ipyparallel as ipp\n",
    "rc = ipp.Client()\n",
    "rc.ids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-11T05:59:18.853907Z",
     "start_time": "2018-09-11T05:59:18.560633Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[stdout:0] Dataset has been loaded\n",
      "[stdout:1] Dataset has been loaded\n",
      "[stdout:2] Dataset has been loaded\n",
      "[stdout:3] Dataset has been loaded\n",
      "[stdout:4] Dataset has been loaded\n",
      "[stdout:5] Dataset has been loaded\n",
      "[stdout:6] Dataset has been loaded\n",
      "[stdout:7] Dataset has been loaded\n"
     ]
    }
   ],
   "source": [
    "%%px\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "import h5py\n",
    "from sklearn.model_selection import train_test_split\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",
    "#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",
    "#create a train - test split of the dataset\n",
    "x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.55, 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",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import hyptune as htun\n",
    "import scnets as scn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-11T06:09:41.335279Z",
     "start_time": "2018-09-11T06:09:41.326581Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AsyncResult: execute>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%px --target 0 --noblock\n",
    "import os\n",
    "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\" \n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"\n",
    "func = scn.resnet\n",
    "param_grid = dict(num_units=[2], \n",
    "                  red_dim=[8,16],\n",
    "                  batch_size=[32, 64],\n",
    "                  ker_size=[3, 5, 7])\n",
    "    \n",
    "cvresults = htun.get_cv_grid(modelfunc=func, \n",
    "                             param_grid=param_grid,\n",
    "                             num_epochs=250,\n",
    "                             x_train=x_train,\n",
    "                             y_train=y_train)\n",
    "\n",
    "htres = htun.print_tuning_results(cvresults, func)\n",
    "htres.to_csv('n2_cvsrch.dat')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-11T06:11:05.592120Z",
     "start_time": "2018-09-11T06:11:05.583939Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AsyncResult: execute>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%px --target 1 --noblock\n",
    "import os\n",
    "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\" \n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"1\"\n",
    "func = scn.resnet\n",
    "param_grid = dict(num_units=[3], \n",
    "                  red_dim=[8,16],\n",
    "                  batch_size=[32, 64],\n",
    "                  ker_size=[3, 5, 7])\n",
    "    \n",
    "cvresults = htun.get_cv_grid(modelfunc=func, \n",
    "                             param_grid=param_grid,\n",
    "                             num_epochs=250,\n",
    "                             x_train=x_train,\n",
    "                             y_train=y_train)\n",
    "\n",
    "htres = htun.print_tuning_results(cvresults, func)\n",
    "htres.to_csv('n3_cvsrch.dat')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-11T06:10:10.156751Z",
     "start_time": "2018-09-11T06:10:10.147263Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AsyncResult: execute>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%px --target 2 --noblock\n",
    "import os\n",
    "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\" \n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"\n",
    "func = scn.resnet\n",
    "param_grid = dict(num_units=[4], \n",
    "                  red_dim=[8,16],\n",
    "                  batch_size=[32, 64],\n",
    "                  ker_size=[3, 5, 7])\n",
    "    \n",
    "cvresults = htun.get_cv_grid(modelfunc=func, \n",
    "                             param_grid=param_grid,\n",
    "                             num_epochs=250,\n",
    "                             x_train=x_train,\n",
    "                             y_train=y_train)\n",
    "\n",
    "htres = htun.print_tuning_results(cvresults, func)\n",
    "htres.to_csv('n4_cvsrch.dat')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-11T06:11:12.169586Z",
     "start_time": "2018-09-11T06:11:12.161932Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AsyncResult: execute>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%px --target 3 --noblock\n",
    "import os\n",
    "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\" \n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"1\"\n",
    "func = scn.resnet\n",
    "param_grid = dict(num_units=[5], \n",
    "                  red_dim=[8,16],\n",
    "                  batch_size=[32, 64],\n",
    "                  ker_size=[3, 5, 7])\n",
    "    \n",
    "cvresults = htun.get_cv_grid(modelfunc=func, \n",
    "                             param_grid=param_grid,\n",
    "                             num_epochs=250,\n",
    "                             x_train=x_train,\n",
    "                             y_train=y_train)\n",
    "\n",
    "htres = htun.print_tuning_results(cvresults, func)\n",
    "htres.to_csv('n5_cvsrch.dat')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-11T06:10:40.811578Z",
     "start_time": "2018-09-11T06:10:40.804963Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AsyncResult: execute>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%px --target 4 --noblock\n",
    "import os\n",
    "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\" \n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"\n",
    "func = scn.resnet\n",
    "param_grid = dict(num_units=[6], \n",
    "                  red_dim=[8,16],\n",
    "                  batch_size=[32, 64],\n",
    "                  ker_size=[3, 5, 7])\n",
    "    \n",
    "cvresults = htun.get_cv_grid(modelfunc=func, \n",
    "                             param_grid=param_grid,\n",
    "                             num_epochs=250,\n",
    "                             x_train=x_train,\n",
    "                             y_train=y_train)\n",
    "\n",
    "htres = htun.print_tuning_results(cvresults, func)\n",
    "htres.to_csv('n6_cvsrch.dat')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}