model_2-0.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n",
      "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/matplotlib/__init__.py:962: UserWarning: Duplicate key in file \"/home/ubuntu/.config/matplotlib/matplotlibrc\", line #2\n",
      "  (fname, cnt))\n",
      "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/matplotlib/__init__.py:962: UserWarning: Duplicate key in file \"/home/ubuntu/.config/matplotlib/matplotlibrc\", line #3\n",
      "  (fname, cnt))\n"
     ]
    }
   ],
   "source": [
    "import model_generator as mg\n",
    "model = mg.Model([300, 400, 40, 1], 'Model_2-0',  r_discount=.95, c_discount=.99)\n",
    "mg.saveHP(model)\n",
    "model.setData()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "num minibatches:3662\n",
      "\tEpoch time:37 seconds\n",
      "Cost after epoch 0: 0.774336\n",
      "\tDev error (no dropout): 0.692408\n",
      "Cost after epoch 1: 0.690412\n",
      "\tDev error (no dropout): 0.678869\n",
      "Cost after epoch 2: 0.676701\n",
      "\tDev error (no dropout): 0.670645\n",
      "Cost after epoch 3: 0.668544\n",
      "\tDev error (no dropout): 0.665694\n",
      "Cost after epoch 4: 0.662730\n",
      "\tDev error (no dropout): 0.661543\n",
      "Cost after epoch 5: 0.658313\n",
      "\tDev error (no dropout): 0.658748\n",
      "Cost after epoch 6: 0.654607\n",
      "\tDev error (no dropout): 0.658014\n",
      "Cost after epoch 7: 0.651704\n",
      "\tDev error (no dropout): 0.655487\n",
      "Cost after epoch 8: 0.649158\n",
      "\tDev error (no dropout): 0.655867\n",
      "Cost after epoch 9: 0.647246\n",
      "\tDev error (no dropout): 0.653992\n",
      "\tAve epoch time: 36.203566\n",
      "\t100 epochs will take 1:00:20.356634\n",
      "Cost after epoch 10: 0.645720\n",
      "\tDev error (no dropout): 0.652714\n",
      "Cost after epoch 20: 0.634486\n",
      "\tDev error (no dropout): 0.652391\n",
      "Cost after epoch 30: 0.627892\n",
      "\tDev error (no dropout): 0.651705\n",
      "Cost after epoch 40: 0.623709\n",
      "\tDev error (no dropout): 0.652934\n",
      "Parameters have been trained!\n",
      "Train cost (No dropout): 0.60834557\n",
      "Test cost (No dropout): 0.6536268\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f28579f4a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.TrainModel(num_epochs=50, learning_rate=0.0001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VS Greedy: 49.250%\tVS Random: 87.360%\n",
      "Standard metric: 641.48\n"
     ]
    }
   ],
   "source": [
    "model.TestModel(10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Simulating...\n",
      "Converting...\n",
      "Saving...\n",
      "done.\n"
     ]
    }
   ],
   "source": [
    "model.GenerateTrainingData()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Environment (conda_tensorflow_p36)",
   "language": "python",
   "name": "conda_tensorflow_p36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}