House_pricing (regression).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.datasets import boston_housing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "(train_data,train_targets),(test_data,test_targets) = boston_housing.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2\n",
      "(404, 13)\n",
      "float64\n",
      "[ 15.2  42.3  50.   21.1  17.7  18.5  11.3  15.6  15.6  14.4  12.1  17.9\n",
      "  23.1  19.9  15.7   8.8  50.   22.5  24.1  27.5  10.9  30.8  32.9  24.\n",
      "  18.5  13.3  22.9  34.7  16.6  17.5  22.3  16.1  14.9  23.1  34.9  25.\n",
      "  13.9  13.1  20.4  20.   15.2  24.7  22.2  16.7  12.7  15.6  18.4  21.\n",
      "  30.1  15.1  18.7   9.6  31.5  24.8  19.1  22.   14.5  11.   32.   29.4\n",
      "  20.3  24.4  14.6  19.5  14.1  14.3  15.6  10.5   6.3  19.3  19.3  13.4\n",
      "  36.4  17.8  13.5  16.5   8.3  14.3  16.   13.4  28.6  43.5  20.2  22.\n",
      "  23.   20.7  12.5  48.5  14.6  13.4  23.7  50.   21.7  39.8  38.7  22.2\n",
      "  34.9  22.5  31.1  28.7  46.   41.7  21.   26.6  15.   24.4  13.3  21.2\n",
      "  11.7  21.7  19.4  50.   22.8  19.7  24.7  36.2  14.2  18.9  18.3  20.6\n",
      "  24.6  18.2   8.7  44.   10.4  13.2  21.2  37.   30.7  22.9  20.   19.3\n",
      "  31.7  32.   23.1  18.8  10.9  50.   19.6   5.   14.4  19.8  13.8  19.6\n",
      "  23.9  24.5  25.   19.9  17.2  24.6  13.5  26.6  21.4  11.9  22.6  19.6\n",
      "   8.5  23.7  23.1  22.4  20.5  23.6  18.4  35.2  23.1  27.9  20.6  23.7\n",
      "  28.   13.6  27.1  23.6  20.6  18.2  21.7  17.1   8.4  25.3  13.8  22.2\n",
      "  18.4  20.7  31.6  30.5  20.3   8.8  19.2  19.4  23.1  23.   14.8  48.8\n",
      "  22.6  33.4  21.1  13.6  32.2  13.1  23.4  18.9  23.9  11.8  23.3  22.8\n",
      "  19.6  16.7  13.4  22.2  20.4  21.8  26.4  14.9  24.1  23.8  12.3  29.1\n",
      "  21.   19.5  23.3  23.8  17.8  11.5  21.7  19.9  25.   33.4  28.5  21.4\n",
      "  24.3  27.5  33.1  16.2  23.3  48.3  22.9  22.8  13.1  12.7  22.6  15.\n",
      "  15.3  10.5  24.   18.5  21.7  19.5  33.2  23.2   5.   19.1  12.7  22.3\n",
      "  10.2  13.9  16.3  17.   20.1  29.9  17.2  37.3  45.4  17.8  23.2  29.\n",
      "  22.   18.   17.4  34.6  20.1  25.   15.6  24.8  28.2  21.2  21.4  23.8\n",
      "  31.   26.2  17.4  37.9  17.5  20.    8.3  23.9   8.4  13.8   7.2  11.7\n",
      "  17.1  21.6  50.   16.1  20.4  20.6  21.4  20.6  36.5   8.5  24.8  10.8\n",
      "  21.9  17.3  18.9  36.2  14.9  18.2  33.3  21.8  19.7  31.6  24.8  19.4\n",
      "  22.8   7.5  44.8  16.8  18.7  50.   50.   19.5  20.1  50.   17.2  20.8\n",
      "  19.3  41.3  20.4  20.5  13.8  16.5  23.9  20.6  31.5  23.3  16.8  14.\n",
      "  33.8  36.1  12.8  18.3  18.7  19.1  29.   30.1  50.   50.   22.   11.9\n",
      "  37.6  50.   22.7  20.8  23.5  27.9  50.   19.3  23.9  22.6  15.2  21.7\n",
      "  19.2  43.8  20.3  33.2  19.9  22.5  32.7  22.   17.1  19.   15.   16.1\n",
      "  25.1  23.7  28.7  37.2  22.6  16.4  25.   29.8  22.1  17.4  18.1  30.3\n",
      "  17.5  24.7  12.6  26.5  28.7  13.3  10.4  24.4  23.   20.   17.8   7.\n",
      "  11.8  24.4  13.8  19.4  25.2  19.4  19.4  29.1]\n",
      "[   1.23247    0.         8.14       0.         0.538      6.142     91.7\n",
      "    3.9769     4.       307.        21.       396.9       18.72   ]\n"
     ]
    }
   ],
   "source": [
    "print(train_data.ndim)\n",
    "print(train_data.shape)\n",
    "print(train_data.dtype)\n",
    "print(train_targets)\n",
    "print(train_data[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean = train_data.mean(axis=0)\n",
    "train_data -= mean\n",
    "std = train_data.std(axis=0)\n",
    "train_data /= std\n",
    "\n",
    "test_data -= mean\n",
    "test_data /= std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import models,layers\n",
    "\n",
    "def build_model():\n",
    "    model = models.Sequential()\n",
    "    \n",
    "    model.add(layers.Dense(64, activation='relu', input_shape=(train_data.shape[1],)))\n",
    "    model.add(layers.Dense(64, activation='relu'))\n",
    "    model.add(layers.Dense(1))\n",
    "    model.compile(optimizer=\"rmsprop\", loss=\"mse\", metrics=['mae'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "k = 4\n",
    "num_val_samples = len(train_data)//4\n",
    "num_epochs = 500"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "processing fold #  1\n",
      "processing fold #  2\n",
      "processing fold #  3\n",
      "processing fold #  4\n",
      "K fold validation finished\n"
     ]
    }
   ],
   "source": [
    "all_scores = []\n",
    "\n",
    "for i in range(k):\n",
    "    print('processing fold # ',i+1)\n",
    "    val_data = train_data[i*num_val_samples:(i+1)*num_val_samples]\n",
    "    val_targets = train_targets[i*num_val_samples:(i+1)*num_val_samples]\n",
    "    \n",
    "    partial_train_data = np.concatenate([train_data[:i*num_val_samples], train_data[(i+1)*num_val_samples:]], \n",
    "                                        axis=0)\n",
    "    partial_train_targets = np.concatenate([train_targets[:i*num_val_samples],\n",
    "                                          train_targets[(i+1)*num_val_samples:]], \n",
    "                                           axis=0)\n",
    "    \n",
    "    model = build_model()\n",
    "    history = model.fit(partial_train_data, partial_train_targets, epochs=num_epochs, batch_size=1, verbose=0, \n",
    "                       validation_data=(val_data,val_targets))\n",
    "    mae_history = history.history['val_mean_absolute_error']\n",
    "    all_scores.append(mae_history)\n",
    "    \n",
    "print(\"K fold validation finished\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_mae_histories = [\n",
    "    np.mean([x[i] for x in all_scores]) for i in range(num_epochs)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'MAE')"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe0dcb203c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def smooth(points, factor=0.9):\n",
    "    smp = []\n",
    "    for point in points:\n",
    "        if smp:\n",
    "            prev = smp[-1]\n",
    "            smp.append(prev*factor + point*(1-factor))\n",
    "        else:\n",
    "            smp.append(point)\n",
    "\n",
    "    return smp\n",
    "\n",
    "avg_mae_histories = smooth(avg_mae_histories[12:])\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.plot(range(1, len(avg_mae_histories)+1), avg_mae_histories)\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('MAE')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "102/102 [==============================] - 0s 2ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[16.492712731454887, 2.5588560758852492]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = build_model()\n",
    "model.fit(train_data, train_targets, epochs=80, batch_size=16, verbose=0)\n",
    "model.evaluate(test_data,test_targets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}