Urban_nn_model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WMlxrqXBFqW0"
   },
   "source": [
    "# Urban Sound Classification - NN Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Bx1c6_z5FqW4"
   },
   "source": [
    "## Fetch Processed Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Y3AZ44NR6I9H"
   },
   "outputs": [],
   "source": [
    "#extracting data from pkl files\n",
    "import pickle as pkl\n",
    "\n",
    "with open(\"x_data.pickle\",\"rb\") as f:\n",
    "  x_data = pkl.load(f)\n",
    "\n",
    "with open(\"y_label.pickle\",\"rb\") as f:\n",
    "  y_label = pkl.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "F5qDzT0aGh8g"
   },
   "outputs": [],
   "source": [
    "#reshape data\n",
    "import numpy as np\n",
    "from keras.utils.np_utils import to_categorical\n",
    "\n",
    "X = []\n",
    "Y = []\n",
    "for i in range(10):\n",
    "  xi = np.array(x_data[i])\n",
    "  xi_2d = np.reshape(xi,(xi.shape[0],xi.shape[1]*xi.shape[2]))\n",
    "  X.append(xi_2d)\n",
    "\n",
    "  yi = to_categorical(y_label[i], num_classes=10)\n",
    "  Y.append(yi)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "x6bzJu5M6dn8",
    "outputId": "97cd42bf-9c1b-4e37-de60-46f914a85cdb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((873, 180), (873, 10), (888, 180), (888, 10))"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#shape\n",
    "X[0].shape, Y[0].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SkVJ82R4FqW9"
   },
   "source": [
    "## Build Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Bj6Y44j1eKbe"
   },
   "outputs": [],
   "source": [
    "from keras import Sequential\n",
    "from keras.layers import Dense,Dropout,Activation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "oHWYBiziNguQ"
   },
   "outputs": [],
   "source": [
    "def get_model():\n",
    "  #forming model\n",
    "  model = Sequential()\n",
    "  \n",
    "  #building the model\n",
    "  model.add(Dense(units=256,activation='relu',input_dim=180))\n",
    "  model.add(Dropout(0.4))\n",
    "  model.add(Dense(units=256,activation='relu'))\n",
    "  model.add(Dropout(0.4))\n",
    "  model.add(Dense(units=256,activation='relu'))\n",
    "  model.add(Dropout(0.4))\n",
    "  model.add(Dense(units=10,activation='softmax'))\n",
    "  \n",
    "  #compiling\n",
    "  model.compile(optimizer='adam',loss='categorical_crossentropy',metrics=['accuracy'])\n",
    "\n",
    "  return model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "n35YI3A8FqXB"
   },
   "source": [
    "## 10-fold Cross Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "PJAmI2AU9XqL",
    "outputId": "7527e995-28ff-4e59-b17e-215df03eb1de"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 5.7872 - accuracy: 0.1703\n",
      "Epoch 2/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 2.2249 - accuracy: 0.2659\n",
      "Epoch 3/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.8882 - accuracy: 0.3450\n",
      "Epoch 4/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.6579 - accuracy: 0.4298\n",
      "Epoch 5/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.5138 - accuracy: 0.4871\n",
      "Epoch 6/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.3916 - accuracy: 0.5298\n",
      "Epoch 7/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.2738 - accuracy: 0.5759\n",
      "Epoch 8/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.1803 - accuracy: 0.6063\n",
      "Epoch 9/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.0950 - accuracy: 0.6360\n",
      "Epoch 10/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.0492 - accuracy: 0.6549\n",
      "Epoch 11/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.9432 - accuracy: 0.6923\n",
      "Epoch 12/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.9364 - accuracy: 0.6923\n",
      "Epoch 13/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.8300 - accuracy: 0.7294\n",
      "Epoch 14/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.8317 - accuracy: 0.7266\n",
      "Epoch 15/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.7687 - accuracy: 0.7464\n",
      "Epoch 16/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.7454 - accuracy: 0.7501\n",
      "Epoch 17/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.6954 - accuracy: 0.7736\n",
      "Epoch 18/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.6767 - accuracy: 0.7810\n",
      "Epoch 19/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.6464 - accuracy: 0.7871\n",
      "Epoch 20/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.6141 - accuracy: 0.7973\n",
      "Epoch 21/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.5997 - accuracy: 0.8068\n",
      "Epoch 22/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.5984 - accuracy: 0.8095\n",
      "Epoch 23/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5605 - accuracy: 0.8146\n",
      "Epoch 24/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.5504 - accuracy: 0.8192\n",
      "Epoch 25/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5325 - accuracy: 0.8280\n",
      "Epoch 26/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5150 - accuracy: 0.8338\n",
      "Epoch 27/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.4985 - accuracy: 0.8346\n",
      "Epoch 28/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.4945 - accuracy: 0.8403\n",
      "Epoch 29/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.4843 - accuracy: 0.8406\n",
      "Epoch 30/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.4723 - accuracy: 0.8451\n",
      "246/246 [==============================] - 1s 2ms/step - loss: 0.2314 - accuracy: 0.9289\n",
      "28/28 [==============================] - 0s 3ms/step - loss: 2.3409 - accuracy: 0.5349\n",
      "Results for folder:  0\n",
      "[0.23141370713710785, 0.9288713335990906]\n",
      "[2.3408572673797607, 0.5349370241165161]\n",
      "Epoch 1/30\n",
      "157/157 [==============================] - 1s 3ms/step - loss: 5.1728 - accuracy: 0.1721\n",
      "Epoch 2/30\n",
      "157/157 [==============================] - 1s 3ms/step - loss: 2.2073 - accuracy: 0.2489\n",
      "Epoch 3/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.9272 - accuracy: 0.3389\n",
      "Epoch 4/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.7169 - accuracy: 0.4131\n",
      "Epoch 5/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.5414 - accuracy: 0.4723\n",
      "Epoch 6/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.3844 - accuracy: 0.5348\n",
      "Epoch 7/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.2567 - accuracy: 0.5816\n",
      "Epoch 8/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.1536 - accuracy: 0.6193\n",
      "Epoch 9/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.0975 - accuracy: 0.6419\n",
      "Epoch 10/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.0143 - accuracy: 0.6669\n",
      "Epoch 11/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.9329 - accuracy: 0.6874\n",
      "Epoch 12/30\n",
      "157/157 [==============================] - 1s 4ms/step - loss: 0.9039 - accuracy: 0.7042\n",
      "Epoch 13/30\n",
      "157/157 [==============================] - 1s 4ms/step - loss: 0.8498 - accuracy: 0.7227\n",
      "Epoch 14/30\n",
      "157/157 [==============================] - 1s 4ms/step - loss: 0.7949 - accuracy: 0.7353\n",
      "Epoch 15/30\n",
      "157/157 [==============================] - 1s 4ms/step - loss: 0.7428 - accuracy: 0.7533\n",
      "Epoch 16/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.7243 - accuracy: 0.7650\n",
      "Epoch 17/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.6792 - accuracy: 0.7761\n",
      "Epoch 18/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.6686 - accuracy: 0.7812\n",
      "Epoch 19/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.6415 - accuracy: 0.7922\n",
      "Epoch 20/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.6274 - accuracy: 0.7950\n",
      "Epoch 21/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5862 - accuracy: 0.8092\n",
      "Epoch 22/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5589 - accuracy: 0.8149\n",
      "Epoch 23/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5573 - accuracy: 0.8182\n",
      "Epoch 24/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5362 - accuracy: 0.8264\n",
      "Epoch 25/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5320 - accuracy: 0.8283\n",
      "Epoch 26/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.4957 - accuracy: 0.8320\n",
      "Epoch 27/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.4949 - accuracy: 0.8354\n",
      "Epoch 28/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.4802 - accuracy: 0.8417\n",
      "Epoch 29/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.4534 - accuracy: 0.8545\n",
      "Epoch 30/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.4726 - accuracy: 0.8445\n",
      "246/246 [==============================] - 1s 2ms/step - loss: 0.2262 - accuracy: 0.9314\n",
      "28/28 [==============================] - 0s 2ms/step - loss: 2.4815 - accuracy: 0.5135\n",
      "Results for folder:  1\n",
      "[0.22622905671596527, 0.9314125180244446]\n",
      "[2.4815266132354736, 0.5135135054588318]\n",
      "Epoch 1/30\n",
      "157/157 [==============================] - 1s 3ms/step - loss: 5.3868 - accuracy: 0.1748\n",
      "Epoch 2/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 2.1566 - accuracy: 0.2815\n",
      "Epoch 3/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.8647 - accuracy: 0.3595\n",
      "Epoch 4/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.6447 - accuracy: 0.4378\n",
      "Epoch 5/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.4933 - accuracy: 0.4916\n",
      "Epoch 6/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.3975 - accuracy: 0.5355\n",
      "Epoch 7/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.2404 - accuracy: 0.5859\n",
      "Epoch 8/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.1514 - accuracy: 0.6152\n",
      "Epoch 9/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.0680 - accuracy: 0.6511\n",
      "Epoch 10/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 1.0015 - accuracy: 0.6720\n",
      "Epoch 11/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.9322 - accuracy: 0.6982\n",
      "Epoch 12/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.8703 - accuracy: 0.7159\n",
      "Epoch 13/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.8365 - accuracy: 0.7245\n",
      "Epoch 14/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.7794 - accuracy: 0.7432\n",
      "Epoch 15/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.7566 - accuracy: 0.7603\n",
      "Epoch 16/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.7536 - accuracy: 0.7619\n",
      "Epoch 17/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.6817 - accuracy: 0.7822\n",
      "Epoch 18/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.6655 - accuracy: 0.7833\n",
      "Epoch 19/30\n",
      "157/157 [==============================] - 1s 3ms/step - loss: 0.6358 - accuracy: 0.7956\n",
      "Epoch 20/30\n",
      "157/157 [==============================] - 1s 3ms/step - loss: 0.5867 - accuracy: 0.8036\n",
      "Epoch 21/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5862 - accuracy: 0.8102\n",
      "Epoch 22/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5829 - accuracy: 0.8115\n",
      "Epoch 23/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5531 - accuracy: 0.8246\n",
      "Epoch 24/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5167 - accuracy: 0.8323\n",
      "Epoch 25/30\n",
      "157/157 [==============================] - 1s 4ms/step - loss: 0.5178 - accuracy: 0.8268\n",
      "Epoch 26/30\n",
      "157/157 [==============================] - 1s 4ms/step - loss: 0.5027 - accuracy: 0.8357\n",
      "Epoch 27/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.4920 - accuracy: 0.8389\n",
      "Epoch 28/30\n",
      "157/157 [==============================] - 1s 4ms/step - loss: 0.4910 - accuracy: 0.8436\n",
      "Epoch 29/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.5101 - accuracy: 0.8360\n",
      "Epoch 30/30\n",
      "157/157 [==============================] - 0s 3ms/step - loss: 0.4659 - accuracy: 0.8503\n",
      "244/244 [==============================] - 1s 2ms/step - loss: 0.2321 - accuracy: 0.9188\n",
      "29/29 [==============================] - 0s 3ms/step - loss: 2.3853 - accuracy: 0.5330\n",
      "Results for folder:  2\n",
      "[0.23209334909915924, 0.9187908172607422]\n",
      "[2.385305643081665, 0.5329729914665222]\n",
      "Epoch 1/30\n",
      "155/155 [==============================] - 1s 3ms/step - loss: 5.3041 - accuracy: 0.1811\n",
      "Epoch 2/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 2.1765 - accuracy: 0.2648\n",
      "Epoch 3/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 1.8472 - accuracy: 0.3568\n",
      "Epoch 4/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 1.6605 - accuracy: 0.4301\n",
      "Epoch 5/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 1.4528 - accuracy: 0.5010\n",
      "Epoch 6/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 1.3273 - accuracy: 0.5469\n",
      "Epoch 7/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 1.2225 - accuracy: 0.5845\n",
      "Epoch 8/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 1.0871 - accuracy: 0.6342\n",
      "Epoch 9/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 1.0415 - accuracy: 0.6502\n",
      "Epoch 10/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.9678 - accuracy: 0.6785\n",
      "Epoch 11/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.9133 - accuracy: 0.7001\n",
      "Epoch 12/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.8630 - accuracy: 0.7184\n",
      "Epoch 13/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.8057 - accuracy: 0.7406\n",
      "Epoch 14/30\n",
      "155/155 [==============================] - 1s 3ms/step - loss: 0.7571 - accuracy: 0.7550\n",
      "Epoch 15/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.7219 - accuracy: 0.7676\n",
      "Epoch 16/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.7005 - accuracy: 0.7754\n",
      "Epoch 17/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.6699 - accuracy: 0.7808\n",
      "Epoch 18/30\n",
      "155/155 [==============================] - 1s 3ms/step - loss: 0.6343 - accuracy: 0.7937\n",
      "Epoch 19/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.6202 - accuracy: 0.8034\n",
      "Epoch 20/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.5762 - accuracy: 0.8065\n",
      "Epoch 21/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.5798 - accuracy: 0.8112\n",
      "Epoch 22/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.5467 - accuracy: 0.8229\n",
      "Epoch 23/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.5524 - accuracy: 0.8201\n",
      "Epoch 24/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.5230 - accuracy: 0.8277\n",
      "Epoch 25/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.4944 - accuracy: 0.8424\n",
      "Epoch 26/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.5011 - accuracy: 0.8369\n",
      "Epoch 27/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.4978 - accuracy: 0.8331\n",
      "Epoch 28/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.4552 - accuracy: 0.8498\n",
      "Epoch 29/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.4568 - accuracy: 0.8489\n",
      "Epoch 30/30\n",
      "155/155 [==============================] - 0s 3ms/step - loss: 0.4557 - accuracy: 0.8515\n",
      "242/242 [==============================] - 1s 2ms/step - loss: 0.2256 - accuracy: 0.9332\n",
      "31/31 [==============================] - 0s 2ms/step - loss: 1.5363 - accuracy: 0.5667\n",
      "Results for folder:  3\n",
      "[0.22556760907173157, 0.9332213997840881]\n",
      "[1.5363050699234009, 0.5666666626930237]\n",
      "Epoch 1/30\n",
      "156/156 [==============================] - 1s 3ms/step - loss: 5.2500 - accuracy: 0.1838\n",
      "Epoch 2/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 2.1248 - accuracy: 0.2741\n",
      "Epoch 3/30\n",
      "156/156 [==============================] - 1s 3ms/step - loss: 1.8813 - accuracy: 0.3485\n",
      "Epoch 4/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 1.7030 - accuracy: 0.4088\n",
      "Epoch 5/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 1.5167 - accuracy: 0.4699\n",
      "Epoch 6/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 1.3886 - accuracy: 0.5250\n",
      "Epoch 7/30\n",
      "156/156 [==============================] - 1s 3ms/step - loss: 1.2928 - accuracy: 0.5616\n",
      "Epoch 8/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 1.1804 - accuracy: 0.6011\n",
      "Epoch 9/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 1.0925 - accuracy: 0.6339\n",
      "Epoch 10/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 1.0004 - accuracy: 0.6626\n",
      "Epoch 11/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.9546 - accuracy: 0.6868\n",
      "Epoch 12/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.8979 - accuracy: 0.7075\n",
      "Epoch 13/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.8507 - accuracy: 0.7178\n",
      "Epoch 14/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.7991 - accuracy: 0.7424\n",
      "Epoch 15/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.7433 - accuracy: 0.7545\n",
      "Epoch 16/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.7261 - accuracy: 0.7609\n",
      "Epoch 17/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.6805 - accuracy: 0.7789\n",
      "Epoch 18/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.6433 - accuracy: 0.7925\n",
      "Epoch 19/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.6314 - accuracy: 0.7944\n",
      "Epoch 20/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.6295 - accuracy: 0.7955\n",
      "Epoch 21/30\n",
      "156/156 [==============================] - 1s 3ms/step - loss: 0.5886 - accuracy: 0.8085\n",
      "Epoch 22/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.5499 - accuracy: 0.8203\n",
      "Epoch 23/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.5344 - accuracy: 0.8204\n",
      "Epoch 24/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.5332 - accuracy: 0.8235\n",
      "Epoch 25/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.5145 - accuracy: 0.8271\n",
      "Epoch 26/30\n",
      "156/156 [==============================] - 0s 3ms/step - loss: 0.5018 - accuracy: 0.8347\n",
      "Epoch 27/30\n",
      "156/156 [==============================] - 1s 3ms/step - loss: 0.4922 - accuracy: 0.8334\n",
      "Epoch 28/30\n",
      "156/156 [==============================] - 1s 4ms/step - loss: 0.4757 - accuracy: 0.8406\n",
      "Epoch 29/30\n",
      "156/156 [==============================] - 1s 4ms/step - loss: 0.4719 - accuracy: 0.8443\n",
      "Epoch 30/30\n",
      "156/156 [==============================] - 1s 4ms/step - loss: 0.4474 - accuracy: 0.8497\n",
      "244/244 [==============================] - 1s 2ms/step - loss: 0.2233 - accuracy: 0.9275\n",
      "30/30 [==============================] - 0s 2ms/step - loss: 1.3059 - accuracy: 0.6453\n",
      "Results for folder:  4\n",
      "[0.22330932319164276, 0.9275269508361816]\n",
      "[1.3059213161468506, 0.6452991366386414]\n",
      "Epoch 1/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 5.1856 - accuracy: 0.1739\n",
      "Epoch 2/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 2.1231 - accuracy: 0.2774\n",
      "Epoch 3/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.8728 - accuracy: 0.3471\n",
      "Epoch 4/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.6786 - accuracy: 0.4253\n",
      "Epoch 5/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.5199 - accuracy: 0.4829\n",
      "Epoch 6/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.3928 - accuracy: 0.5327\n",
      "Epoch 7/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.2615 - accuracy: 0.5800\n",
      "Epoch 8/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.1796 - accuracy: 0.6126\n",
      "Epoch 9/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.0890 - accuracy: 0.6381\n",
      "Epoch 10/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.0310 - accuracy: 0.6630\n",
      "Epoch 11/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.9367 - accuracy: 0.6931\n",
      "Epoch 12/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.8984 - accuracy: 0.7038\n",
      "Epoch 13/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.8756 - accuracy: 0.7145\n",
      "Epoch 14/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.8176 - accuracy: 0.7318\n",
      "Epoch 15/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.7759 - accuracy: 0.7447\n",
      "Epoch 16/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.7328 - accuracy: 0.7605\n",
      "Epoch 17/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.6905 - accuracy: 0.7734\n",
      "Epoch 18/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.6765 - accuracy: 0.7791\n",
      "Epoch 19/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.6473 - accuracy: 0.7890\n",
      "Epoch 20/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.6031 - accuracy: 0.8007\n",
      "Epoch 21/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5951 - accuracy: 0.8069\n",
      "Epoch 22/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5690 - accuracy: 0.8133\n",
      "Epoch 23/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5510 - accuracy: 0.8136\n",
      "Epoch 24/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5449 - accuracy: 0.8174\n",
      "Epoch 25/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5310 - accuracy: 0.8278\n",
      "Epoch 26/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5282 - accuracy: 0.8322\n",
      "Epoch 27/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.4930 - accuracy: 0.8371\n",
      "Epoch 28/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.4895 - accuracy: 0.8373\n",
      "Epoch 29/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.4926 - accuracy: 0.8380\n",
      "Epoch 30/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.4503 - accuracy: 0.8530\n",
      "248/248 [==============================] - 1s 2ms/step - loss: 0.2348 - accuracy: 0.9270\n",
      "26/26 [==============================] - 0s 2ms/step - loss: 2.2027 - accuracy: 0.5103\n",
      "Results for folder:  5\n",
      "[0.23480136692523956, 0.9270451664924622]\n",
      "[2.2027130126953125, 0.5103280544281006]\n",
      "Epoch 1/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 5.0996 - accuracy: 0.1903\n",
      "Epoch 2/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 2.1681 - accuracy: 0.2868\n",
      "Epoch 3/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.8346 - accuracy: 0.3675\n",
      "Epoch 4/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.6111 - accuracy: 0.4500\n",
      "Epoch 5/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 1.4631 - accuracy: 0.4972\n",
      "Epoch 6/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.3332 - accuracy: 0.5470\n",
      "Epoch 7/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 1.2211 - accuracy: 0.5868\n",
      "Epoch 8/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.1602 - accuracy: 0.6121\n",
      "Epoch 9/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.0467 - accuracy: 0.6529\n",
      "Epoch 10/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.9776 - accuracy: 0.6754\n",
      "Epoch 11/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.9453 - accuracy: 0.6870\n",
      "Epoch 12/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.8681 - accuracy: 0.7131\n",
      "Epoch 13/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.8258 - accuracy: 0.7302\n",
      "Epoch 14/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.7669 - accuracy: 0.7451\n",
      "Epoch 15/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.7312 - accuracy: 0.7568\n",
      "Epoch 16/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.7008 - accuracy: 0.7715\n",
      "Epoch 17/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.6785 - accuracy: 0.7722\n",
      "Epoch 18/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.6471 - accuracy: 0.7916\n",
      "Epoch 19/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.6340 - accuracy: 0.7941\n",
      "Epoch 20/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.6038 - accuracy: 0.8049\n",
      "Epoch 21/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5838 - accuracy: 0.8111\n",
      "Epoch 22/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.5524 - accuracy: 0.8185\n",
      "Epoch 23/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5413 - accuracy: 0.8156\n",
      "Epoch 24/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.5264 - accuracy: 0.8285\n",
      "Epoch 25/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5030 - accuracy: 0.8322\n",
      "Epoch 26/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.5100 - accuracy: 0.8313\n",
      "Epoch 27/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5036 - accuracy: 0.8338\n",
      "Epoch 28/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.4837 - accuracy: 0.8395\n",
      "Epoch 29/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.4733 - accuracy: 0.8386\n",
      "Epoch 30/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.4519 - accuracy: 0.8517\n",
      "247/247 [==============================] - 1s 2ms/step - loss: 0.2104 - accuracy: 0.9335\n",
      "27/27 [==============================] - 0s 2ms/step - loss: 1.4456 - accuracy: 0.5835\n",
      "Results for folder:  6\n",
      "[0.2104264199733734, 0.9334937930107117]\n",
      "[1.4455633163452148, 0.5835322141647339]\n",
      "Epoch 1/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 5.0975 - accuracy: 0.1755\n",
      "Epoch 2/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 2.1831 - accuracy: 0.2653\n",
      "Epoch 3/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.9065 - accuracy: 0.3313\n",
      "Epoch 4/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.6988 - accuracy: 0.4090\n",
      "Epoch 5/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.5280 - accuracy: 0.4815\n",
      "Epoch 6/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.3664 - accuracy: 0.5305\n",
      "Epoch 7/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.2374 - accuracy: 0.5695\n",
      "Epoch 8/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.1311 - accuracy: 0.6241\n",
      "Epoch 9/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 1.0599 - accuracy: 0.6441\n",
      "Epoch 10/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.9911 - accuracy: 0.6752\n",
      "Epoch 11/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.9252 - accuracy: 0.6948\n",
      "Epoch 12/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.8662 - accuracy: 0.7140\n",
      "Epoch 13/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.8204 - accuracy: 0.7371\n",
      "Epoch 14/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.7849 - accuracy: 0.7444\n",
      "Epoch 15/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.7412 - accuracy: 0.7615\n",
      "Epoch 16/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.6858 - accuracy: 0.7725\n",
      "Epoch 17/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.6868 - accuracy: 0.7732\n",
      "Epoch 18/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.6489 - accuracy: 0.7897\n",
      "Epoch 19/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.6209 - accuracy: 0.8005\n",
      "Epoch 20/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5845 - accuracy: 0.8100\n",
      "Epoch 21/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5988 - accuracy: 0.8029\n",
      "Epoch 22/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5609 - accuracy: 0.8190\n",
      "Epoch 23/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5497 - accuracy: 0.8197\n",
      "Epoch 24/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5180 - accuracy: 0.8304\n",
      "Epoch 25/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.4929 - accuracy: 0.8386\n",
      "Epoch 26/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.4911 - accuracy: 0.8388\n",
      "Epoch 27/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.4958 - accuracy: 0.8354\n",
      "Epoch 28/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.4879 - accuracy: 0.8375\n",
      "Epoch 29/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.4559 - accuracy: 0.8491\n",
      "Epoch 30/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.4781 - accuracy: 0.8463\n",
      "248/248 [==============================] - 1s 3ms/step - loss: 0.2337 - accuracy: 0.9243\n",
      "26/26 [==============================] - 0s 3ms/step - loss: 1.6315 - accuracy: 0.6079\n",
      "Results for folder:  7\n",
      "[0.23373307287693024, 0.9242997765541077]\n",
      "[1.6315255165100098, 0.6079404354095459]\n",
      "Epoch 1/30\n",
      "159/159 [==============================] - 1s 4ms/step - loss: 5.2464 - accuracy: 0.1753\n",
      "Epoch 2/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 2.1939 - accuracy: 0.2520\n",
      "Epoch 3/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.9407 - accuracy: 0.3301\n",
      "Epoch 4/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.7285 - accuracy: 0.3999\n",
      "Epoch 5/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.5756 - accuracy: 0.4625\n",
      "Epoch 6/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.4305 - accuracy: 0.5200\n",
      "Epoch 7/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.3019 - accuracy: 0.5595\n",
      "Epoch 8/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.2129 - accuracy: 0.5987\n",
      "Epoch 9/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.1257 - accuracy: 0.6282\n",
      "Epoch 10/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 1.0487 - accuracy: 0.6541\n",
      "Epoch 11/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.9648 - accuracy: 0.6809\n",
      "Epoch 12/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.9325 - accuracy: 0.6971\n",
      "Epoch 13/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.8674 - accuracy: 0.7117\n",
      "Epoch 14/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.8202 - accuracy: 0.7302\n",
      "Epoch 15/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.7700 - accuracy: 0.7521\n",
      "Epoch 16/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.7568 - accuracy: 0.7535\n",
      "Epoch 17/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.7161 - accuracy: 0.7689\n",
      "Epoch 18/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.6810 - accuracy: 0.7785\n",
      "Epoch 19/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.6455 - accuracy: 0.7869\n",
      "Epoch 20/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.6482 - accuracy: 0.7928\n",
      "Epoch 21/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5990 - accuracy: 0.7993\n",
      "Epoch 22/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5979 - accuracy: 0.8063\n",
      "Epoch 23/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5666 - accuracy: 0.8125\n",
      "Epoch 24/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.5449 - accuracy: 0.8210\n",
      "Epoch 25/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5727 - accuracy: 0.8180\n",
      "Epoch 26/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5209 - accuracy: 0.8278\n",
      "Epoch 27/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5060 - accuracy: 0.8363\n",
      "Epoch 28/30\n",
      "159/159 [==============================] - 0s 3ms/step - loss: 0.4841 - accuracy: 0.8422\n",
      "Epoch 29/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.4889 - accuracy: 0.8353\n",
      "Epoch 30/30\n",
      "159/159 [==============================] - 1s 3ms/step - loss: 0.5067 - accuracy: 0.8418\n",
      "248/248 [==============================] - 1s 2ms/step - loss: 0.2617 - accuracy: 0.9181\n",
      "26/26 [==============================] - 0s 3ms/step - loss: 1.9105 - accuracy: 0.6176\n",
      "Results for folder:  8\n",
      "[0.26168516278266907, 0.9181404709815979]\n",
      "[1.9104586839675903, 0.6176470518112183]\n",
      "Epoch 1/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 5.1000 - accuracy: 0.1801\n",
      "Epoch 2/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 2.1198 - accuracy: 0.2863\n",
      "Epoch 3/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 1.8164 - accuracy: 0.3731\n",
      "Epoch 4/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.5999 - accuracy: 0.4497\n",
      "Epoch 5/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 1.4164 - accuracy: 0.5192\n",
      "Epoch 6/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 1.2931 - accuracy: 0.5601\n",
      "Epoch 7/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.1932 - accuracy: 0.6015\n",
      "Epoch 8/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 1.1148 - accuracy: 0.6284\n",
      "Epoch 9/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 1.0259 - accuracy: 0.6612\n",
      "Epoch 10/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.9619 - accuracy: 0.6849\n",
      "Epoch 11/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.9288 - accuracy: 0.6975\n",
      "Epoch 12/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.8704 - accuracy: 0.7234\n",
      "Epoch 13/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.8223 - accuracy: 0.7363\n",
      "Epoch 14/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.7461 - accuracy: 0.7536\n",
      "Epoch 15/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.7422 - accuracy: 0.7562\n",
      "Epoch 16/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.7234 - accuracy: 0.7663\n",
      "Epoch 17/30\n",
      "158/158 [==============================] - 0s 3ms/step - loss: 0.6827 - accuracy: 0.7780\n",
      "Epoch 18/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.6602 - accuracy: 0.7849\n",
      "Epoch 19/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.6374 - accuracy: 0.7924\n",
      "Epoch 20/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.6001 - accuracy: 0.8052\n",
      "Epoch 21/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5887 - accuracy: 0.8056\n",
      "Epoch 22/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5634 - accuracy: 0.8141\n",
      "Epoch 23/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5702 - accuracy: 0.8130\n",
      "Epoch 24/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5571 - accuracy: 0.8223\n",
      "Epoch 25/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5179 - accuracy: 0.8315\n",
      "Epoch 26/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5185 - accuracy: 0.8313\n",
      "Epoch 27/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.5050 - accuracy: 0.8358\n",
      "Epoch 28/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.4659 - accuracy: 0.8502\n",
      "Epoch 29/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.4781 - accuracy: 0.8457\n",
      "Epoch 30/30\n",
      "158/158 [==============================] - 1s 3ms/step - loss: 0.4533 - accuracy: 0.8517\n",
      "247/247 [==============================] - 1s 2ms/step - loss: 0.2239 - accuracy: 0.9310\n",
      "27/27 [==============================] - 0s 2ms/step - loss: 1.2169 - accuracy: 0.6308\n",
      "Results for folder:  9\n",
      "[0.2238582968711853, 0.9309689402580261]\n",
      "[1.2168772220611572, 0.6308243870735168]\n"
     ]
    }
   ],
   "source": [
    "sum_train_loss = 0\n",
    "sum_test_loss = 0\n",
    "sum_train_acc = 0\n",
    "sum_test_acc = 0\n",
    "\n",
    "for i in range(10):\n",
    "  model = get_model()\n",
    "  \n",
    "  x_train = np.concatenate([X[j] for j in [k for k in range(10) if k not in [i]]])\n",
    "  y_train = np.concatenate([Y[j] for j in [k for k in range(10) if k not in [i]]])\n",
    "\n",
    "  x_test = X[i]\n",
    "  y_test = Y[i]\n",
    "\n",
    "  model.fit(x_train, y_train, epochs=30, batch_size=50)\n",
    "\n",
    "  #train and test loss and scores respectively\n",
    "  train_eval = model.evaluate(x_train, y_train)\n",
    "  test_eval = model.evaluate(x_test, y_test)\n",
    "\n",
    "  print(\"Results for folder: \", i)\n",
    "  print(train_eval)\n",
    "  print(test_eval)\n",
    "\n",
    "  sum_train_loss += train_eval[0]\n",
    "  sum_test_loss += test_eval[0]\n",
    "  sum_train_acc += train_eval[1]\n",
    "  sum_test_acc += test_eval[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "85wBQth3FqXB"
   },
   "source": [
    "## Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "EngIWFCdATNP",
    "outputId": "1a821a56-aa22-4ad5-d03a-bfe01f5151a2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Avg Train Loss:  0.23031173646450043\n",
      "Avg Train Acc:  0.9273771166801452\n",
      "Avg Test Loss Score:  1.8457053661346436\n",
      "Avg Test Loss Acc:  0.5743661463260651\n"
     ]
    }
   ],
   "source": [
    "print(\"Avg Train Loss: \", sum_train_loss/10)\n",
    "print(\"Avg Train Acc: \", sum_train_acc/10)\n",
    "print(\"Avg Test Loss Score: \", sum_test_loss/10)\n",
    "print(\"Avg Test Loss Acc: \", sum_test_acc/10)"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "provenance": []
  },
  "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}