4_Scalar Regression-Boston housing data.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Scalar Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Design a neural network to predict the median prices of homes in a give Boston suburb in the mid-1970's, using the Python library Keras."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.4.3'"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import keras\n",
    "keras.__version__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## About the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Boston data frame has 506 rows and 14 columns.\n",
    "\n",
    "This data frame contains the following columns:\n",
    "\n",
    "`crim` : \n",
    "per capita crime rate by town.\n",
    "\n",
    "`zn` : \n",
    "proportion of residential land zoned for lots over 25,000 sq.ft.\n",
    "\n",
    "`indus` : \n",
    "proportion of non-retail business acres per town.\n",
    "\n",
    "`chas` : \n",
    "Charles River dummy variable (= 1 if tract bounds river; 0 otherwise).\n",
    "\n",
    "`nox` : \n",
    "nitrogen oxides concentration (parts per 10 million).\n",
    "\n",
    "`rm` : \n",
    "average number of rooms per dwelling.\n",
    "\n",
    "`age` : \n",
    "proportion of owner-occupied units built prior to 1940.\n",
    "\n",
    "`dis` : \n",
    "weighted mean of distances to five Boston employment centres.\n",
    "\n",
    "`rad` : \n",
    "index of accessibility to radial highways.\n",
    "\n",
    "`tax` : \n",
    "full-value property-tax rate per \\$10,000.\n",
    "\n",
    "`ptratio` : \n",
    "pupil-teacher ratio by town.\n",
    "\n",
    "`black` : \n",
    "1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town.\n",
    "\n",
    "`lstat` : \n",
    "lower status of the population (percent).\n",
    "\n",
    "`medv` : \n",
    "median value of owner-occupied homes in \\$1000s.\n",
    "\n",
    "Source\n",
    "Harrison, D. and Rubinfeld, D.L. (1978) Hedonic prices and the demand for clean air. J. Environ. Economics and Management 5, 81–102.\n",
    "\n",
    "Belsley D.A., Kuh, E. and Welsch, R.E. (1980) Regression Diagnostics. Identifying Influential Data and Sources of Collinearity. New York: Wiley.\n",
    "\n",
    "It has relatively few data points: only 506, split\n",
    "between 404 training samples and 102 test samples. And each feature in the input data (for example, the crime rate) has a different scale."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/boston_housing.npz\n",
      "57344/57026 [==============================] - 0s 5us/step\n"
     ]
    }
   ],
   "source": [
    "from keras.datasets import boston_housing\n",
    "\n",
    "(train_data, train_targets),(test_data, test_targets) = boston_housing.load_data() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(404, 13)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(102, 13)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([15.2, 42.3, 50. , 21.1, 17.7, 18.5, 11.3, 15.6, 15.6, 14.4, 12.1,\n",
       "       17.9, 23.1, 19.9, 15.7,  8.8, 50. , 22.5, 24.1, 27.5, 10.9, 30.8,\n",
       "       32.9, 24. , 18.5, 13.3, 22.9, 34.7, 16.6, 17.5, 22.3, 16.1, 14.9,\n",
       "       23.1, 34.9, 25. , 13.9, 13.1, 20.4, 20. , 15.2, 24.7, 22.2, 16.7,\n",
       "       12.7, 15.6, 18.4, 21. , 30.1, 15.1, 18.7,  9.6, 31.5, 24.8, 19.1,\n",
       "       22. , 14.5, 11. , 32. , 29.4, 20.3, 24.4, 14.6, 19.5, 14.1, 14.3,\n",
       "       15.6, 10.5,  6.3, 19.3, 19.3, 13.4, 36.4, 17.8, 13.5, 16.5,  8.3,\n",
       "       14.3, 16. , 13.4, 28.6, 43.5, 20.2, 22. , 23. , 20.7, 12.5, 48.5,\n",
       "       14.6, 13.4, 23.7, 50. , 21.7, 39.8, 38.7, 22.2, 34.9, 22.5, 31.1,\n",
       "       28.7, 46. , 41.7, 21. , 26.6, 15. , 24.4, 13.3, 21.2, 11.7, 21.7,\n",
       "       19.4, 50. , 22.8, 19.7, 24.7, 36.2, 14.2, 18.9, 18.3, 20.6, 24.6,\n",
       "       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,\n",
       "       19.6, 23.9, 24.5, 25. , 19.9, 17.2, 24.6, 13.5, 26.6, 21.4, 11.9,\n",
       "       22.6, 19.6,  8.5, 23.7, 23.1, 22.4, 20.5, 23.6, 18.4, 35.2, 23.1,\n",
       "       27.9, 20.6, 23.7, 28. , 13.6, 27.1, 23.6, 20.6, 18.2, 21.7, 17.1,\n",
       "        8.4, 25.3, 13.8, 22.2, 18.4, 20.7, 31.6, 30.5, 20.3,  8.8, 19.2,\n",
       "       19.4, 23.1, 23. , 14.8, 48.8, 22.6, 33.4, 21.1, 13.6, 32.2, 13.1,\n",
       "       23.4, 18.9, 23.9, 11.8, 23.3, 22.8, 19.6, 16.7, 13.4, 22.2, 20.4,\n",
       "       21.8, 26.4, 14.9, 24.1, 23.8, 12.3, 29.1, 21. , 19.5, 23.3, 23.8,\n",
       "       17.8, 11.5, 21.7, 19.9, 25. , 33.4, 28.5, 21.4, 24.3, 27.5, 33.1,\n",
       "       16.2, 23.3, 48.3, 22.9, 22.8, 13.1, 12.7, 22.6, 15. , 15.3, 10.5,\n",
       "       24. , 18.5, 21.7, 19.5, 33.2, 23.2,  5. , 19.1, 12.7, 22.3, 10.2,\n",
       "       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,\n",
       "       23.8, 31. , 26.2, 17.4, 37.9, 17.5, 20. ,  8.3, 23.9,  8.4, 13.8,\n",
       "        7.2, 11.7, 17.1, 21.6, 50. , 16.1, 20.4, 20.6, 21.4, 20.6, 36.5,\n",
       "        8.5, 24.8, 10.8, 21.9, 17.3, 18.9, 36.2, 14.9, 18.2, 33.3, 21.8,\n",
       "       19.7, 31.6, 24.8, 19.4, 22.8,  7.5, 44.8, 16.8, 18.7, 50. , 50. ,\n",
       "       19.5, 20.1, 50. , 17.2, 20.8, 19.3, 41.3, 20.4, 20.5, 13.8, 16.5,\n",
       "       23.9, 20.6, 31.5, 23.3, 16.8, 14. , 33.8, 36.1, 12.8, 18.3, 18.7,\n",
       "       19.1, 29. , 30.1, 50. , 50. , 22. , 11.9, 37.6, 50. , 22.7, 20.8,\n",
       "       23.5, 27.9, 50. , 19.3, 23.9, 22.6, 15.2, 21.7, 19.2, 43.8, 20.3,\n",
       "       33.2, 19.9, 22.5, 32.7, 22. , 17.1, 19. , 15. , 16.1, 25.1, 23.7,\n",
       "       28.7, 37.2, 22.6, 16.4, 25. , 29.8, 22.1, 17.4, 18.1, 30.3, 17.5,\n",
       "       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])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_targets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Targets are the median values of owner-occupied homes, in thousands of dollars."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preparing the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Every feature of the dataset takes on very different ranges. While a neural netwok will certainly adapt to such _hetrogeneous_ data, it would make learning more difficult. To deal with this we shall implement feature-wise normalization: _for each feature in the input data, subtract the mean of the feature and divide by standard deviation, so that the feature is centered around 0 with unit standard deviation_. \n",
    "\n",
    "We will do this using Numpy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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",
    "# Normalizing the testing data\n",
    "test_data -= mean\n",
    "test_data /= std"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building our network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import models\n",
    "from keras import layers\n",
    "\n",
    "def build_model():\n",
    "    model = models.Sequential()\n",
    "    model.add(layers.Dense(64, \n",
    "                           activation='relu', \n",
    "                           input_shape=(train_data.shape[1],)))\n",
    "    model.add(layers.Dense(64, activation='relu'))\n",
    "    model.add(layers.Dense(1))\n",
    "    # Compile model\n",
    "    model.compile(optimizer='rmsprop', \n",
    "                  loss='mse', \n",
    "                  metrics=['mae'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setting up K-fold validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "k = 4\n",
    "num_val_samples = len(train_data)//k\n",
    "num_epochs = 100\n",
    "all_scores = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "processing fold # 0\n",
      "Epoch 1/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 202.1898 - mae: 10.5561\n",
      "Epoch 2/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 27.8429 - mae: 3.6085\n",
      "Epoch 3/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 21.3226 - mae: 3.0266\n",
      "Epoch 4/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 18.8393 - mae: 2.8397\n",
      "Epoch 5/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 16.6630 - mae: 2.7011\n",
      "Epoch 6/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 15.2541 - mae: 2.5479\n",
      "Epoch 7/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 14.3217 - mae: 2.5050\n",
      "Epoch 8/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 13.2853 - mae: 2.3510\n",
      "Epoch 9/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 13.4549 - mae: 2.3941\n",
      "Epoch 10/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 12.8623 - mae: 2.3662\n",
      "Epoch 11/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.8497 - mae: 2.2954\n",
      "Epoch 12/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.6276 - mae: 2.2709\n",
      "Epoch 13/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.3640 - mae: 2.2173\n",
      "Epoch 14/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.0167 - mae: 2.2298\n",
      "Epoch 15/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.3329 - mae: 2.1537\n",
      "Epoch 16/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.8186 - mae: 2.2046\n",
      "Epoch 17/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 10.1575 - mae: 2.1372\n",
      "Epoch 18/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.9122 - mae: 2.0141\n",
      "Epoch 19/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 9.5883 - mae: 2.0855\n",
      "Epoch 20/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.3596 - mae: 2.0490\n",
      "Epoch 21/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 8.7744 - mae: 1.9981\n",
      "Epoch 22/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.0171 - mae: 2.0031\n",
      "Epoch 23/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.5426 - mae: 2.0377\n",
      "Epoch 24/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.6847 - mae: 2.0242\n",
      "Epoch 25/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.7462 - mae: 1.9747\n",
      "Epoch 26/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.5036 - mae: 1.9199\n",
      "Epoch 27/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.8920 - mae: 1.8842\n",
      "Epoch 28/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 8.3225 - mae: 1.9477\n",
      "Epoch 29/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.8056 - mae: 1.9003\n",
      "Epoch 30/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.5106 - mae: 1.8814\n",
      "Epoch 31/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.7228 - mae: 1.8392\n",
      "Epoch 32/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.5783 - mae: 1.8676\n",
      "Epoch 33/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.1540 - mae: 1.8100\n",
      "Epoch 34/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.4659 - mae: 1.8251\n",
      "Epoch 35/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.3177 - mae: 1.8199\n",
      "Epoch 36/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.0252 - mae: 1.8035\n",
      "Epoch 37/100\n",
      "303/303 [==============================] - 2s 5ms/step - loss: 6.9785 - mae: 1.7515\n",
      "Epoch 38/100\n",
      "303/303 [==============================] - 2s 6ms/step - loss: 6.7553 - mae: 1.7320\n",
      "Epoch 39/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 6.2007 - mae: 1.7391\n",
      "Epoch 40/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.9069 - mae: 1.7595\n",
      "Epoch 41/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.6607 - mae: 1.7303\n",
      "Epoch 42/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.4284 - mae: 1.6855\n",
      "Epoch 43/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.9030 - mae: 1.7797\n",
      "Epoch 44/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.3527 - mae: 1.6567\n",
      "Epoch 45/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.4932 - mae: 1.6799\n",
      "Epoch 46/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.2544 - mae: 1.6787\n",
      "Epoch 47/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.4261 - mae: 1.7144\n",
      "Epoch 48/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.0470 - mae: 1.6904\n",
      "Epoch 49/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 6.0138 - mae: 1.6501\n",
      "Epoch 50/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.2058 - mae: 1.6436\n",
      "Epoch 51/100\n",
      "303/303 [==============================] - 2s 6ms/step - loss: 5.8783 - mae: 1.6626\n",
      "Epoch 52/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.0168 - mae: 1.6854\n",
      "Epoch 53/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.2644 - mae: 1.6455\n",
      "Epoch 54/100\n",
      "303/303 [==============================] - 2s 5ms/step - loss: 5.6063 - mae: 1.5804\n",
      "Epoch 55/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.8832 - mae: 1.6384\n",
      "Epoch 56/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.0770 - mae: 1.6262\n",
      "Epoch 57/100\n",
      "303/303 [==============================] - 2s 5ms/step - loss: 5.5939 - mae: 1.6082\n",
      "Epoch 58/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.3412 - mae: 1.5020\n",
      "Epoch 59/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.5544 - mae: 1.5844\n",
      "Epoch 60/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.6696 - mae: 1.6109\n",
      "Epoch 61/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.7117 - mae: 1.5594\n",
      "Epoch 62/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.4843 - mae: 1.5777\n",
      "Epoch 63/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 5.4037 - mae: 1.6297\n",
      "Epoch 64/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.6279 - mae: 1.5986\n",
      "Epoch 65/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.5329 - mae: 1.6029\n",
      "Epoch 66/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.3513 - mae: 1.5535\n",
      "Epoch 67/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 5.6699 - mae: 1.6007\n",
      "Epoch 68/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.3057 - mae: 1.5100\n",
      "Epoch 69/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 5.4794 - mae: 1.5339\n",
      "Epoch 70/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.3531 - mae: 1.5548\n",
      "Epoch 71/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.2151 - mae: 1.5600\n",
      "Epoch 72/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 5.3061 - mae: 1.5795\n",
      "Epoch 73/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.2294 - mae: 1.5909\n",
      "Epoch 74/100\n",
      "303/303 [==============================] - 2s 5ms/step - loss: 5.1198 - mae: 1.5048\n",
      "Epoch 75/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 4.9780 - mae: 1.4720\n",
      "Epoch 76/100\n",
      "303/303 [==============================] - 2s 5ms/step - loss: 5.1012 - mae: 1.5076\n",
      "Epoch 77/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.9918 - mae: 1.4954\n",
      "Epoch 78/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.7539 - mae: 1.4802\n",
      "Epoch 79/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.4854 - mae: 1.4418\n",
      "Epoch 80/100\n",
      "303/303 [==============================] - 2s 5ms/step - loss: 4.8472 - mae: 1.4838\n",
      "Epoch 81/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.7258 - mae: 1.4624\n",
      "Epoch 82/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.9854 - mae: 1.5244\n",
      "Epoch 83/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8493 - mae: 1.5013\n",
      "Epoch 84/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.8183 - mae: 1.4823\n",
      "Epoch 85/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "303/303 [==============================] - 1s 3ms/step - loss: 4.4491 - mae: 1.4562\n",
      "Epoch 86/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.7790 - mae: 1.4533\n",
      "Epoch 87/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.6931 - mae: 1.4421\n",
      "Epoch 88/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6514 - mae: 1.4524\n",
      "Epoch 89/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8054 - mae: 1.4628\n",
      "Epoch 90/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.7762 - mae: 1.4202\n",
      "Epoch 91/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1626 - mae: 1.3784\n",
      "Epoch 92/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5848 - mae: 1.4238\n",
      "Epoch 93/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.4589 - mae: 1.4258\n",
      "Epoch 94/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3133 - mae: 1.4375\n",
      "Epoch 95/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0754 - mae: 1.3400\n",
      "Epoch 96/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.2478 - mae: 1.4076\n",
      "Epoch 97/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3645 - mae: 1.4037\n",
      "Epoch 98/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5491 - mae: 1.4619\n",
      "Epoch 99/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.2166 - mae: 1.3736\n",
      "Epoch 100/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1153 - mae: 1.4001\n",
      "4/4 [==============================] - 0s 8ms/step - loss: 10.8632 - mae: 2.3696\n",
      "processing fold # 1\n",
      "Epoch 1/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 198.9538 - mae: 10.4273\n",
      "Epoch 2/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 31.6141 - mae: 3.7648\n",
      "Epoch 3/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 23.4288 - mae: 3.2244\n",
      "Epoch 4/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 19.6661 - mae: 2.9782\n",
      "Epoch 5/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 17.4010 - mae: 2.7423\n",
      "Epoch 6/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 15.6809 - mae: 2.5752\n",
      "Epoch 7/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 14.4297 - mae: 2.4496\n",
      "Epoch 8/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 13.4466 - mae: 2.3879\n",
      "Epoch 9/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 12.7304 - mae: 2.2640\n",
      "Epoch 10/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 12.2877 - mae: 2.2597\n",
      "Epoch 11/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.8592 - mae: 2.2159\n",
      "Epoch 12/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.2040 - mae: 2.2184\n",
      "Epoch 13/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.0346 - mae: 2.1691\n",
      "Epoch 14/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.8567 - mae: 2.1097\n",
      "Epoch 15/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.5241 - mae: 2.0866\n",
      "Epoch 16/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.6243 - mae: 2.0994\n",
      "Epoch 17/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.2025 - mae: 2.0732\n",
      "Epoch 18/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.6317 - mae: 2.0557\n",
      "Epoch 19/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.9651 - mae: 1.9858\n",
      "Epoch 20/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 9.2484 - mae: 1.9902\n",
      "Epoch 21/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.2694 - mae: 2.0172\n",
      "Epoch 22/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.7370 - mae: 1.9446\n",
      "Epoch 23/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 8.9830 - mae: 1.8905\n",
      "Epoch 24/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 8.7001 - mae: 1.9131\n",
      "Epoch 25/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 9.0661 - mae: 1.9553\n",
      "Epoch 26/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.5372 - mae: 1.8748\n",
      "Epoch 27/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 8.3633 - mae: 1.8943\n",
      "Epoch 28/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.6768 - mae: 1.8539\n",
      "Epoch 29/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.2384 - mae: 1.8268\n",
      "Epoch 30/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 8.3049 - mae: 1.8043\n",
      "Epoch 31/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.7490 - mae: 1.9279\n",
      "Epoch 32/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.9197 - mae: 1.7934\n",
      "Epoch 33/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.4123 - mae: 1.8296\n",
      "Epoch 34/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 7.5534 - mae: 1.7358\n",
      "Epoch 35/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.7604 - mae: 1.7992\n",
      "Epoch 36/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.6331 - mae: 1.7554\n",
      "Epoch 37/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.3639 - mae: 1.7651\n",
      "Epoch 38/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.2094 - mae: 1.7541\n",
      "Epoch 39/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.2316 - mae: 1.7516\n",
      "Epoch 40/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.1838 - mae: 1.6754\n",
      "Epoch 41/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.7736 - mae: 1.7394\n",
      "Epoch 42/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.7518 - mae: 1.7249\n",
      "Epoch 43/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.5780 - mae: 1.7009\n",
      "Epoch 44/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 6.8532 - mae: 1.6526\n",
      "Epoch 45/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.8969 - mae: 1.6176\n",
      "Epoch 46/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.5251 - mae: 1.6535\n",
      "Epoch 47/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.8605 - mae: 1.6380\n",
      "Epoch 48/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.6245 - mae: 1.6264\n",
      "Epoch 49/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.2172 - mae: 1.6347\n",
      "Epoch 50/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.9474 - mae: 1.5775\n",
      "Epoch 51/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.4394 - mae: 1.5622\n",
      "Epoch 52/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.1054 - mae: 1.5605\n",
      "Epoch 53/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.0946 - mae: 1.6056\n",
      "Epoch 54/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.1199 - mae: 1.6058\n",
      "Epoch 55/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.8257 - mae: 1.5499\n",
      "Epoch 56/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.8097 - mae: 1.5368\n",
      "Epoch 57/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.5126 - mae: 1.5723\n",
      "Epoch 58/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.5660 - mae: 1.4716\n",
      "Epoch 59/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.7602 - mae: 1.5603\n",
      "Epoch 60/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.4357 - mae: 1.4724\n",
      "Epoch 61/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.4806 - mae: 1.5633\n",
      "Epoch 62/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.1747 - mae: 1.4806\n",
      "Epoch 63/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.2990 - mae: 1.5160\n",
      "Epoch 64/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.1029 - mae: 1.4729\n",
      "Epoch 65/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3410 - mae: 1.4444\n",
      "Epoch 66/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.2973 - mae: 1.4801\n",
      "Epoch 67/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.9201 - mae: 1.4206\n",
      "Epoch 68/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "303/303 [==============================] - 1s 3ms/step - loss: 5.0907 - mae: 1.4420\n",
      "Epoch 69/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 5.1045 - mae: 1.4448\n",
      "Epoch 70/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.0449 - mae: 1.4037\n",
      "Epoch 71/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5920 - mae: 1.3821\n",
      "Epoch 72/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.4879 - mae: 1.3692\n",
      "Epoch 73/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8344 - mae: 1.4204\n",
      "Epoch 74/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8368 - mae: 1.3959\n",
      "Epoch 75/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6180 - mae: 1.3799\n",
      "Epoch 76/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.3334 - mae: 1.3897\n",
      "Epoch 77/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5625 - mae: 1.4043\n",
      "Epoch 78/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3316 - mae: 1.2746\n",
      "Epoch 79/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3799 - mae: 1.3564\n",
      "Epoch 80/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.2399 - mae: 1.2967\n",
      "Epoch 81/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5044 - mae: 1.3597\n",
      "Epoch 82/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.2102 - mae: 1.3334\n",
      "Epoch 83/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5058 - mae: 1.3845\n",
      "Epoch 84/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1186 - mae: 1.3081\n",
      "Epoch 85/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0624 - mae: 1.3671A: 0s - loss: 2.720\n",
      "Epoch 86/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1275 - mae: 1.3343\n",
      "Epoch 87/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3069 - mae: 1.2808\n",
      "Epoch 88/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.9502 - mae: 1.3399\n",
      "Epoch 89/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0669 - mae: 1.3376\n",
      "Epoch 90/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1222 - mae: 1.3170\n",
      "Epoch 91/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0080 - mae: 1.2725\n",
      "Epoch 92/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.8834 - mae: 1.3194\n",
      "Epoch 93/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.7621 - mae: 1.3375\n",
      "Epoch 94/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.8656 - mae: 1.2362\n",
      "Epoch 95/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.5799 - mae: 1.2303\n",
      "Epoch 96/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.5178 - mae: 1.2146\n",
      "Epoch 97/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.7122 - mae: 1.2393\n",
      "Epoch 98/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.7498 - mae: 1.2532\n",
      "Epoch 99/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.8314 - mae: 1.2633\n",
      "Epoch 100/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.4353 - mae: 1.2326\n",
      "4/4 [==============================] - 0s 7ms/step - loss: 17.4398 - mae: 2.9285\n",
      "processing fold # 2\n",
      "Epoch 1/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 168.8595 - mae: 9.3210\n",
      "Epoch 2/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 27.4459 - mae: 3.4907\n",
      "Epoch 3/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 19.8341 - mae: 2.9247\n",
      "Epoch 4/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 16.6706 - mae: 2.6945\n",
      "Epoch 5/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 14.5703 - mae: 2.6084\n",
      "Epoch 6/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 13.2007 - mae: 2.4386\n",
      "Epoch 7/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 11.8022 - mae: 2.3578\n",
      "Epoch 8/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.8225 - mae: 2.3633\n",
      "Epoch 9/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.6934 - mae: 2.2483\n",
      "Epoch 10/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.2363 - mae: 2.1323\n",
      "Epoch 11/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 9.9631 - mae: 2.1775\n",
      "Epoch 12/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 9.7178 - mae: 2.1636\n",
      "Epoch 13/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.4801 - mae: 2.1286\n",
      "Epoch 14/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.8350 - mae: 2.0892\n",
      "Epoch 15/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.5096 - mae: 2.0204\n",
      "Epoch 16/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.4279 - mae: 1.9741\n",
      "Epoch 17/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 8.4823 - mae: 2.0176\n",
      "Epoch 18/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.1205 - mae: 1.9830\n",
      "Epoch 19/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.7183 - mae: 1.9127\n",
      "Epoch 20/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.0102 - mae: 1.9462\n",
      "Epoch 21/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.7673 - mae: 1.8810\n",
      "Epoch 22/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.0662 - mae: 1.9182\n",
      "Epoch 23/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.4450 - mae: 1.8527\n",
      "Epoch 24/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.1318 - mae: 1.8024\n",
      "Epoch 25/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.3211 - mae: 1.8662\n",
      "Epoch 26/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.9600 - mae: 1.8617\n",
      "Epoch 27/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.6764 - mae: 1.7857\n",
      "Epoch 28/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.2207 - mae: 1.7602\n",
      "Epoch 29/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.4870 - mae: 1.7570\n",
      "Epoch 30/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.0783 - mae: 1.7769\n",
      "Epoch 31/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.2248 - mae: 1.6815\n",
      "Epoch 32/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.0231 - mae: 1.7347\n",
      "Epoch 33/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.7643 - mae: 1.7015\n",
      "Epoch 34/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.1947 - mae: 1.6561\n",
      "Epoch 35/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.3206 - mae: 1.6857\n",
      "Epoch 36/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.8860 - mae: 1.6798\n",
      "Epoch 37/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.6900 - mae: 1.6809\n",
      "Epoch 38/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.6770 - mae: 1.6111\n",
      "Epoch 39/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3977 - mae: 1.6008\n",
      "Epoch 40/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3554 - mae: 1.6135\n",
      "Epoch 41/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.5453 - mae: 1.5857\n",
      "Epoch 42/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.5007 - mae: 1.5841\n",
      "Epoch 43/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.2786 - mae: 1.5966\n",
      "Epoch 44/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.1333 - mae: 1.5842\n",
      "Epoch 45/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8103 - mae: 1.4878\n",
      "Epoch 46/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3704 - mae: 1.5647\n",
      "Epoch 47/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.2813 - mae: 1.5779\n",
      "Epoch 48/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6741 - mae: 1.5418\n",
      "Epoch 49/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.9722 - mae: 1.5344\n",
      "Epoch 50/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8104 - mae: 1.5119\n",
      "Epoch 51/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "303/303 [==============================] - 1s 3ms/step - loss: 4.7379 - mae: 1.5168\n",
      "Epoch 52/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6552 - mae: 1.4706\n",
      "Epoch 53/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3342 - mae: 1.4434\n",
      "Epoch 54/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.7480 - mae: 1.4844\n",
      "Epoch 55/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3798 - mae: 1.4261\n",
      "Epoch 56/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.4377 - mae: 1.5105\n",
      "Epoch 57/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.0750 - mae: 1.4053\n",
      "Epoch 58/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5303 - mae: 1.4784\n",
      "Epoch 59/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.2979 - mae: 1.4496\n",
      "Epoch 60/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.1547 - mae: 1.3697\n",
      "Epoch 61/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1554 - mae: 1.4306\n",
      "Epoch 62/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1576 - mae: 1.4354\n",
      "Epoch 63/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.2556 - mae: 1.3827\n",
      "Epoch 64/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5259 - mae: 1.4112\n",
      "Epoch 65/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0481 - mae: 1.4212\n",
      "Epoch 66/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.9657 - mae: 1.3773\n",
      "Epoch 67/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0421 - mae: 1.4014\n",
      "Epoch 68/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.2175 - mae: 1.4114\n",
      "Epoch 69/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0551 - mae: 1.4043\n",
      "Epoch 70/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.8463 - mae: 1.3812\n",
      "Epoch 71/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.6009 - mae: 1.3703\n",
      "Epoch 72/100\n",
      "303/303 [==============================] - 2s 5ms/step - loss: 3.7511 - mae: 1.3373\n",
      "Epoch 73/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.9958 - mae: 1.4181\n",
      "Epoch 74/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.5623 - mae: 1.3657\n",
      "Epoch 75/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.8753 - mae: 1.3838\n",
      "Epoch 76/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.8964 - mae: 1.3438\n",
      "Epoch 77/100\n",
      "303/303 [==============================] - 1s 5ms/step - loss: 3.4775 - mae: 1.3051\n",
      "Epoch 78/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.6204 - mae: 1.3150\n",
      "Epoch 79/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.7994 - mae: 1.3444\n",
      "Epoch 80/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.3544 - mae: 1.2958\n",
      "Epoch 81/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.7557 - mae: 1.3262\n",
      "Epoch 82/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.5000 - mae: 1.2844\n",
      "Epoch 83/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 3.2846 - mae: 1.3107\n",
      "Epoch 84/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.2996 - mae: 1.3077\n",
      "Epoch 85/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.0922 - mae: 1.2773\n",
      "Epoch 86/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 3.3313 - mae: 1.2563\n",
      "Epoch 87/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 3.3866 - mae: 1.3079\n",
      "Epoch 88/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 3.0801 - mae: 1.2310\n",
      "Epoch 89/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.3925 - mae: 1.2696\n",
      "Epoch 90/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.2802 - mae: 1.2400\n",
      "Epoch 91/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.1927 - mae: 1.2778\n",
      "Epoch 92/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.1082 - mae: 1.2388\n",
      "Epoch 93/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.1012 - mae: 1.2813\n",
      "Epoch 94/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.0112 - mae: 1.2359\n",
      "Epoch 95/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 3.1683 - mae: 1.2381\n",
      "Epoch 96/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 2.8902 - mae: 1.1466\n",
      "Epoch 97/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.0154 - mae: 1.2360\n",
      "Epoch 98/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 2.8590 - mae: 1.2195\n",
      "Epoch 99/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 2.9206 - mae: 1.2099\n",
      "Epoch 100/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 2.6108 - mae: 1.1579\n",
      "4/4 [==============================] - 0s 6ms/step - loss: 15.5844 - mae: 2.6196\n",
      "processing fold # 3\n",
      "Epoch 1/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 194.5413 - mae: 10.2544\n",
      "Epoch 2/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 26.0220 - mae: 3.5587\n",
      "Epoch 3/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 19.0902 - mae: 3.0228\n",
      "Epoch 4/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 15.5016 - mae: 2.7396\n",
      "Epoch 5/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 14.2901 - mae: 2.4976\n",
      "Epoch 6/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 12.6593 - mae: 2.4380\n",
      "Epoch 7/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 12.2351 - mae: 2.3337\n",
      "Epoch 8/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 11.2643 - mae: 2.2362\n",
      "Epoch 9/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 10.9448 - mae: 2.2830\n",
      "Epoch 10/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.5826 - mae: 2.1989\n",
      "Epoch 11/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 10.3113 - mae: 2.1168\n",
      "Epoch 12/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 10.1206 - mae: 2.1323\n",
      "Epoch 13/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.3552 - mae: 2.0861\n",
      "Epoch 14/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.3740 - mae: 2.0912\n",
      "Epoch 15/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 9.1192 - mae: 2.0036\n",
      "Epoch 16/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.7242 - mae: 2.0078\n",
      "Epoch 17/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.5963 - mae: 2.0498\n",
      "Epoch 18/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.3159 - mae: 1.9276\n",
      "Epoch 19/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.9133 - mae: 1.9789\n",
      "Epoch 20/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.1133 - mae: 1.9319\n",
      "Epoch 21/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 8.3140 - mae: 1.9819\n",
      "Epoch 22/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.7727 - mae: 1.8837\n",
      "Epoch 23/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 7.9954 - mae: 1.8888\n",
      "Epoch 24/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.7560 - mae: 1.8701\n",
      "Epoch 25/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 7.0079 - mae: 1.8395\n",
      "Epoch 26/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 7.6931 - mae: 1.8994\n",
      "Epoch 27/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.0927 - mae: 1.8117\n",
      "Epoch 28/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 6.9114 - mae: 1.8672\n",
      "Epoch 29/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.5524 - mae: 1.8171\n",
      "Epoch 30/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 7.1894 - mae: 1.8139\n",
      "Epoch 31/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.9966 - mae: 1.7684\n",
      "Epoch 32/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.8361 - mae: 1.7934\n",
      "Epoch 33/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.8497 - mae: 1.6947\n",
      "Epoch 34/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "303/303 [==============================] - 1s 3ms/step - loss: 6.5275 - mae: 1.6709\n",
      "Epoch 35/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.6474 - mae: 1.7187\n",
      "Epoch 36/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.8108 - mae: 1.7260\n",
      "Epoch 37/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.9558 - mae: 1.6881\n",
      "Epoch 38/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 6.4670 - mae: 1.6429\n",
      "Epoch 39/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.2934 - mae: 1.6848\n",
      "Epoch 40/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.5211 - mae: 1.6937\n",
      "Epoch 41/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.4117 - mae: 1.7181\n",
      "Epoch 42/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 6.2863 - mae: 1.6647\n",
      "Epoch 43/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.0392 - mae: 1.5971\n",
      "Epoch 44/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.9287 - mae: 1.6323\n",
      "Epoch 45/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 5.5869 - mae: 1.6078\n",
      "Epoch 46/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.9547 - mae: 1.6468\n",
      "Epoch 47/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 6.0332 - mae: 1.6002\n",
      "Epoch 48/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.4097 - mae: 1.5898\n",
      "Epoch 49/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.9389 - mae: 1.6188\n",
      "Epoch 50/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.9133 - mae: 1.5769\n",
      "Epoch 51/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.7179 - mae: 1.6029\n",
      "Epoch 52/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3988 - mae: 1.5672\n",
      "Epoch 53/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.4770 - mae: 1.5181\n",
      "Epoch 54/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.8057 - mae: 1.6037\n",
      "Epoch 55/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.4424 - mae: 1.5799\n",
      "Epoch 56/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.5535 - mae: 1.5645\n",
      "Epoch 57/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 5.4119 - mae: 1.5247\n",
      "Epoch 58/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3313 - mae: 1.5010\n",
      "Epoch 59/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3641 - mae: 1.5327\n",
      "Epoch 60/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.9007 - mae: 1.4796\n",
      "Epoch 61/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.3489 - mae: 1.5428\n",
      "Epoch 62/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.1184 - mae: 1.5134\n",
      "Epoch 63/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8810 - mae: 1.4939\n",
      "Epoch 64/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8531 - mae: 1.4634\n",
      "Epoch 65/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.1847 - mae: 1.5268\n",
      "Epoch 66/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8366 - mae: 1.4605\n",
      "Epoch 67/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.9193 - mae: 1.5031\n",
      "Epoch 68/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8908 - mae: 1.4465\n",
      "Epoch 69/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 5.0046 - mae: 1.5571\n",
      "Epoch 70/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.8485 - mae: 1.4360\n",
      "Epoch 71/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6635 - mae: 1.4388\n",
      "Epoch 72/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.8091 - mae: 1.4688\n",
      "Epoch 73/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6724 - mae: 1.4997\n",
      "Epoch 74/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6129 - mae: 1.4395\n",
      "Epoch 75/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6266 - mae: 1.4260\n",
      "Epoch 76/100\n",
      "303/303 [==============================] - 1s 4ms/step - loss: 4.3976 - mae: 1.4296\n",
      "Epoch 77/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3784 - mae: 1.4107\n",
      "Epoch 78/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3653 - mae: 1.4186\n",
      "Epoch 79/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.2101 - mae: 1.3945\n",
      "Epoch 80/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5916 - mae: 1.4322\n",
      "Epoch 81/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.5467 - mae: 1.4142\n",
      "Epoch 82/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6188 - mae: 1.3865\n",
      "Epoch 83/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.2637 - mae: 1.3554\n",
      "Epoch 84/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1840 - mae: 1.3675\n",
      "Epoch 85/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.6118 - mae: 1.4134\n",
      "Epoch 86/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.3136 - mae: 1.3436\n",
      "Epoch 87/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.5773 - mae: 1.3868\n",
      "Epoch 88/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.0634 - mae: 1.3003\n",
      "Epoch 89/100\n",
      "303/303 [==============================] - 1s 2ms/step - loss: 4.3636 - mae: 1.3855\n",
      "Epoch 90/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0118 - mae: 1.3361 ETA: 0s - loss: 2.8261 -\n",
      "Epoch 91/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1353 - mae: 1.3582\n",
      "Epoch 92/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0737 - mae: 1.3002\n",
      "Epoch 93/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0833 - mae: 1.3520\n",
      "Epoch 94/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.8982 - mae: 1.3154\n",
      "Epoch 95/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.9399 - mae: 1.3720\n",
      "Epoch 96/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.9151 - mae: 1.3064\n",
      "Epoch 97/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.9711 - mae: 1.3284\n",
      "Epoch 98/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.1748 - mae: 1.3449\n",
      "Epoch 99/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 4.0370 - mae: 1.3130\n",
      "Epoch 100/100\n",
      "303/303 [==============================] - 1s 3ms/step - loss: 3.6587 - mae: 1.2915\n",
      "4/4 [==============================] - 0s 3ms/step - loss: 14.7449 - mae: 2.6698\n"
     ]
    }
   ],
   "source": [
    "for i in range(k):\n",
    "    print('processing fold #', i)\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],\n",
    "                                        train_data[(i+1) * num_val_samples:]],\n",
    "                                       axis=0)\n",
    "    partial_train_target = np.concatenate([train_targets[:i * num_val_samples],\n",
    "                                        train_targets[(i+1) * num_val_samples:]],\n",
    "                                       axis=0)\n",
    "    model = build_model()\n",
    "    model.fit(partial_train_data,\n",
    "              partial_train_target,\n",
    "              epochs=num_epochs,\n",
    "              batch_size = 1,\n",
    "              verbose=1)\n",
    "    val_mse, val_mae = model.evaluate(val_data, val_targets, verbose=1)\n",
    "    all_scores.append(val_mae)"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}