ISS296-Phase1Model(AvgPool-kernel3x3)-TestCopy.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\rassa\\anaconda3\\lib\\site-packages\\tensorflow\\python\\compat\\v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "non-resource variables are not supported in the long term\n",
      "curses is not supported on this machine (please install/reinstall curses for an optimal experience)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import sys\n",
    "import tflearn.data_utils as du\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten, Conv2D, MaxPool2D, AveragePooling2D \n",
    "from keras.optimizers import RMSprop\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from keras.preprocessing.image import load_img \n",
    "from keras.preprocessing.image import img_to_array \n",
    "from keras.preprocessing.image import array_to_img\n",
    "import warnings\n",
    "import matplotlib.image as mpimg\n",
    "import os \n",
    "import cv2\n",
    "import tensorflow as tf \n",
    "from PIL import Image \n",
    "from numpy import asarray\n",
    "from tqdm import tqdm\n",
    "from tkinter import Tk     # from tkinter import Tk for Python 3.x\n",
    "from tkinter.filedialog import askopenfilename\n",
    "from tkinter import messagebox\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████| 13440/13440 [00:13<00:00, 1018.61it/s]\n",
      "100%|████████████████████████████████████████████████████████████████████████████| 3360/3360 [00:01<00:00, 2002.91it/s]\n"
     ]
    }
   ],
   "source": [
    "#we need to convert the images to a matrix so we can build our test and train sets\n",
    "\n",
    "DIRECTORY = r\"C:\\Users\\rassa\\Documents\\Machine Learning A-Z  Udemy\\Arabic Handwritten Characters Dataset\"\n",
    "train_data = []\n",
    "test_data = []\n",
    "trainPath = DIRECTORY + r\"\\Train Images 13440x32x32\\train\"\n",
    "testPath  = DIRECTORY + r\"\\Test Images 3360x32x32\\test\"\n",
    "\n",
    "def getImgId(imgName):\n",
    "    return int(imgName.split('_')[1])\n",
    "\n",
    "def makeData(data, directory, sortingCriteria):\n",
    "    imgFileNameList = os.listdir(directory)     #get list of directories in path (ie. list of image names)\n",
    "    imgFileNameList.sort(key = sortingCriteria) #sort list based on key\n",
    "    \n",
    "    for img in tqdm(imgFileNameList):  # iterate over the sorted list\n",
    "                try:\n",
    "                    img_array = cv2.imread(os.path.join(directory, img) ,cv2.IMREAD_GRAYSCALE)  # convert image to array\n",
    "                    data.append([img_array])  # add this to our data\n",
    "                except Exception as e:  # just in case there is a mistake.\n",
    "                    pass\n",
    "\n",
    "makeData(train_data, trainPath, getImgId)\n",
    "makeData(test_data, testPath, getImgId)\n",
    "\n",
    "# convert to ndarray and reshape\n",
    "train_data = np.asarray(train_data).reshape([-1, 32, 32, 1])\n",
    "test_data = np.asarray(test_data).reshape([-1, 32, 32, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Load the labels\n",
    "#header = 0 because\n",
    "#this creates a pandas dataframe (basically a table)\n",
    "train_label = pd.read_csv(DIRECTORY + '/csvTrainLabel 13440x1.csv', header = None)\n",
    "test_label = pd.read_csv(DIRECTORY +  '/csvTestLabel 3360x1.csv', header = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train_data = train_data.iloc[:,:].values.astype('float32')\n",
    "train_label = train_label.iloc[:,:].values.astype('int32')-1\n",
    "#test_data = test_data.iloc[:,:].values.astype('float32') we don't need these anymore\n",
    "test_label = test_label.iloc[:,:].values.astype('int32')-1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_label = du.to_categorical(train_label,28)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = train_data/255\n",
    "test_data = test_data/255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train_data, mean1 = du.featurewise_zero_center(train_data)\n",
    "#test_data, mean2 = du.featurewise_zero_center(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "recognizer = Sequential()\n",
    "#for any kind of deep learning we use the sequential model in keras then \n",
    "#add layers to it\n",
    "recognizer.add(Conv2D(filters = 32, kernel_size = (3,3),padding = 'Same', \n",
    "                 activation ='relu', input_shape = (32,32,1)))\n",
    "#the first 2 dimensional convolusion layer will have 32 filters. filters or kernels are \n",
    "#what we use to extract features. in this case they are of size 3x3 (kernel size)\n",
    "#in the first layer it is necessary to specify the shape so for our case its a\n",
    "#32x32 pixel image and since it's black and white then it has only one dimension or channel\n",
    "#if it was colored then we would have (32,32,3)\n",
    "#activation='relu' this is rectified linear unit. the output filters or convolved layers\n",
    "#might contain some negative values so we apply the rectifier function (or other functions) to break linearity\n",
    "recognizer.add(Conv2D(filters = 32, kernel_size = (3,3),padding = 'Same', \n",
    "                 activation ='relu'))\n",
    "recognizer.add(AveragePooling2D(pool_size=(2,2)))\n",
    "recognizer.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "recognizer.add(Conv2D(filters = 64, kernel_size = (3,3),padding = 'Same', \n",
    "                 activation ='relu'))\n",
    "recognizer.add(Conv2D(filters = 64, kernel_size = (3,3),padding = 'Same', \n",
    "                 activation ='relu'))\n",
    "recognizer.add(MaxPool2D(pool_size=(2,2), strides=(1,1)))\n",
    "recognizer.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "recognizer.add(Flatten())\n",
    "recognizer.add(Dense(units = 256, input_dim = 1024, activation = 'relu'))\n",
    "recognizer.add(Dense(units = 256, activation = \"relu\"))\n",
    "recognizer.add(Dropout(0.5))\n",
    "recognizer.add(Dense(28, activation = \"softmax\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 32, 32, 32)        320       \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 32, 32, 32)        9248      \n",
      "_________________________________________________________________\n",
      "average_pooling2d (AveragePo (None, 16, 16, 32)        0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 16, 16, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 16, 16, 64)        18496     \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 16, 16, 64)        36928     \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 15, 15, 64)        0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 15, 15, 64)        0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 14400)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 256)               3686656   \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 256)               65792     \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 28)                7196      \n",
      "=================================================================\n",
      "Total params: 3,824,636\n",
      "Trainable params: 3,824,636\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "recognizer.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = RMSprop(lr=0.001, rho=0.9, epsilon=1e-08, decay=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "recognizer.compile(optimizer = optimizer , loss = \"categorical_crossentropy\", metrics=[\"accuracy\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "datagen = ImageDataGenerator(\n",
    "        featurewise_center=False, \n",
    "        samplewise_center=False,  \n",
    "        featurewise_std_normalization=False,\n",
    "        samplewise_std_normalization=False,\n",
    "        zca_whitening=False,\n",
    "        rotation_range=10,\n",
    "        zoom_range = 0.1,  \n",
    "        width_shift_range=0.1, \n",
    "        height_shift_range=0.1,\n",
    "        horizontal_flip=False,\n",
    "        vertical_flip=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "datagen.fit(train_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\rassa\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training_v1.py:1240: UserWarning: `model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
      "  warnings.warn('`model.fit_generator` is deprecated and '\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "134/134 - 31s - loss: 2.3736 - acc: 0.2806\n",
      "Epoch 2/30\n",
      "134/134 - 30s - loss: 1.2867 - acc: 0.5784\n",
      "Epoch 3/30\n",
      "134/134 - 30s - loss: 0.9239 - acc: 0.6936\n",
      "Epoch 4/30\n",
      "134/134 - 31s - loss: 0.7148 - acc: 0.7678\n",
      "Epoch 5/30\n",
      "134/134 - 30s - loss: 0.5792 - acc: 0.8052\n",
      "Epoch 6/30\n",
      "134/134 - 30s - loss: 0.5039 - acc: 0.8356\n",
      "Epoch 7/30\n",
      "134/134 - 30s - loss: 0.4425 - acc: 0.8576\n",
      "Epoch 8/30\n",
      "134/134 - 31s - loss: 0.4107 - acc: 0.8682\n",
      "Epoch 9/30\n",
      "134/134 - 30s - loss: 0.3730 - acc: 0.8802\n",
      "Epoch 10/30\n",
      "134/134 - 31s - loss: 0.3434 - acc: 0.8912\n",
      "Epoch 11/30\n",
      "134/134 - 32s - loss: 0.3095 - acc: 0.9013\n",
      "Epoch 12/30\n",
      "134/134 - 35s - loss: 0.2937 - acc: 0.9078\n",
      "Epoch 13/30\n",
      "134/134 - 33s - loss: 0.2786 - acc: 0.9113\n",
      "Epoch 14/30\n",
      "134/134 - 32s - loss: 0.2635 - acc: 0.9164\n",
      "Epoch 15/30\n",
      "134/134 - 32s - loss: 0.2599 - acc: 0.9196\n",
      "Epoch 16/30\n",
      "134/134 - 32s - loss: 0.2350 - acc: 0.9232\n",
      "Epoch 17/30\n",
      "134/134 - 32s - loss: 0.2386 - acc: 0.9250\n",
      "Epoch 18/30\n",
      "134/134 - 31s - loss: 0.2199 - acc: 0.9298\n",
      "Epoch 19/30\n",
      "134/134 - 32s - loss: 0.2179 - acc: 0.9328\n",
      "Epoch 20/30\n",
      "134/134 - 35s - loss: 0.2092 - acc: 0.9346\n",
      "Epoch 21/30\n",
      "134/134 - 33s - loss: 0.2066 - acc: 0.9354\n",
      "Epoch 22/30\n",
      "134/134 - 36s - loss: 0.1926 - acc: 0.9406\n",
      "Epoch 23/30\n",
      "134/134 - 39s - loss: 0.1876 - acc: 0.9398\n",
      "Epoch 24/30\n",
      "134/134 - 37s - loss: 0.1840 - acc: 0.9418\n",
      "Epoch 25/30\n",
      "134/134 - 37s - loss: 0.1871 - acc: 0.9408\n",
      "Epoch 26/30\n",
      "134/134 - 35s - loss: 0.1731 - acc: 0.9454\n",
      "Epoch 27/30\n",
      "134/134 - 34s - loss: 0.1817 - acc: 0.9436\n",
      "Epoch 28/30\n",
      "134/134 - 37s - loss: 0.1805 - acc: 0.9447\n",
      "Epoch 29/30\n",
      "134/134 - 34s - loss: 0.1765 - acc: 0.9457\n",
      "Epoch 30/30\n",
      "134/134 - 34s - loss: 0.1609 - acc: 0.9499\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x241f6c0e9a0>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recognizer.fit_generator(datagen.flow(train_data,train_label, batch_size=100),\n",
    "                             epochs = 30, verbose = 2, steps_per_epoch=train_data.shape[0] // 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\rassa\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training.py:2325: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n",
      "  warnings.warn('`Model.state_updates` will be removed in a future version. '\n"
     ]
    }
   ],
   "source": [
    "predictions = recognizer.predict(test_data)\n",
    "predictions = np.argmax(predictions,axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm = confusion_matrix(test_label, predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy = 0.9630952380952381\n"
     ]
    }
   ],
   "source": [
    "# accuracy: (tp + tn) / (p + n)\n",
    "accuracy = sum(cm[i][i] for i in range(28)) / test_label.shape[0]\n",
    "print(\"accuracy = \" + str(accuracy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[120   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0 119   0   0   0   0   0   0   0   0   0   1   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0 118   2   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   6 113   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   1   0   0   0]\n",
      " [  0   0   0   0 119   1   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   2 118   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   2 118   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0 118   0   2   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   6 107   1   2   0   0   0   0   0   2   0\n",
      "    0   0   0   0   2   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   1   0 117   1   0   0   0   0   0   0   0\n",
      "    0   0   0   0   1   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   2   2   9 106   0   0   0   0   0   0   0\n",
      "    0   0   0   1   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0 119   0   1   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   0 119   0   1   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   3   0 117   0   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   0   1   5 114   0   0   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   0   0   0   0 117   1   0\n",
      "    0   0   0   0   0   0   0   2   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   4 116   0\n",
      "    0   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   1   1   7   1   0   0   0   0   0   0   0   0   0   0 109\n",
      "    0   0   0   0   0   1   0   0   0   0]\n",
      " [  0   0   0   0   0   0   6   0   0   0   0   0   0   0   0   0   0   0\n",
      "  114   0   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   0   0   1   0   0   0   0\n",
      "    0 114   5   0   0   0   0   0   0   0]\n",
      " [  0   0   1   0   0   0   0   0   0   0   0   0   0   1   0   0   0   0\n",
      "    0   2 116   0   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   1   0   0   0   0   0   0\n",
      "    0   0   0 118   0   0   0   0   1   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   3 117   0   0   0   0   0]\n",
      " [  1   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0 118   0   1   0   0]\n",
      " [  0   0   5   0   0   0   0   0   0   0   0   1   0   0   0   0   0   0\n",
      "    0   1   0   5   0   0 108   0   0   0]\n",
      " [  0   0   0   1   0   0   0   0   0   0   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0 117   2   0]\n",
      " [  0   0   0   0   0   0   0   1   0   1   0   0   0   0   0   0   0   0\n",
      "    0   0   0   0   0   0   0   4 114   0]\n",
      " [  0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   1   0\n",
      "    0   0   2   0   0   0   1   0   0 116]]\n"
     ]
    }
   ],
   "source": [
    "print(cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def outputMsg(l):\n",
    "    switcher = {\n",
    "        0: \"zero\",\n",
    "        1 : \"alif\",\n",
    "        2 : \"ba2\",\n",
    "        3: \"ta2\",\n",
    "        4: \"tha2\",\n",
    "        5: \"jha2\",\n",
    "        6: \"7a2\",\n",
    "        7: \"5a2\",\n",
    "        8: \"da2\",\n",
    "        9: \"thal\",\n",
    "        10: \"ra2\",\n",
    "        11: \"za2\",\n",
    "        12: \"seen\", \n",
    "        13: \"sha2\",\n",
    "        14: \"saad\",\n",
    "        15: \"thaad\",\n",
    "        16: \"ttaa2\",\n",
    "        17: \"ththa\",\n",
    "        18: \"3ayn\",\n",
    "        19: \"ghayn\",\n",
    "        20: \"fa2\",\n",
    "        21: \"9a2\",\n",
    "        22: \"ka2\",\n",
    "        23: \"lam\",\n",
    "        24: \"meem\",\n",
    "        25: \"noon\",\n",
    "        26: \"haa2\",\n",
    "        27: \"waw\",\n",
    "        28: \"ya2\"\n",
    "    }\n",
    "    print(\"Your letter is \" + switcher.get(l, \"Invalid label\"))\n",
    "    #messagebox.showinfo(\"Prediction\", \"Your letter is \" + switcher.get(l, \"Invalid label\"))\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "* Enter the alpha value [1.0-3.0]: 1.25\n",
      "* Enter the beta value [0-100]: 25\n",
      "174.0\n",
      "[[1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
      "  0. 0. 0. 0.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Read image given by user\n",
    "\n",
    "\n",
    "Tk().withdraw() # we don't want a full GUI, so keep the root window from appearing\n",
    "filename = askopenfilename() # show an \"Open\" dialog box and return the path to the selected file\n",
    "image = cv2.imread(filename)#r\"C:\\Users\\rassa\\Desktop\\Desktop2.0\\CNN\\9alb-Ba2-0.png\") #(cv.samples.findFile(args.input))\n",
    "\n",
    "contrast_img = np.zeros(image.shape, image.dtype)\n",
    "\n",
    "alpha = 1.0 # Simple contrast control\n",
    "beta = 0    # Simple brightness control\n",
    "\n",
    "# Initialize values\n",
    "try:\n",
    "    alpha = float(input('* Enter the alpha value [1.0-3.0]: '))\n",
    "    beta = int(input('* Enter the beta value [0-100]: '))\n",
    "except ValueError:\n",
    "    print('Error, not a number')\n",
    "    \n",
    "# Do the operation new_image(i,j) = alpha*image(i,j) + beta\n",
    "# Instead of these 'for' loops we could have used simply:\n",
    "# new_image = cv.convertScaleAbs(image, alpha=alpha, beta=beta)\n",
    "# but we wanted to show you how to access the pixels :)\n",
    "for y in range(image.shape[0]):\n",
    "    for x in range(image.shape[1]):\n",
    "        for c in range(image.shape[2]):\n",
    "            # the clip makes sure that the new values of the pixel are between 0 and 255\n",
    "            contrast_img[y,x,c] = np.clip(alpha*image[y,x,c] + beta, 0, 255)\n",
    "        \n",
    "#removethe three dimensions and turn image to gray scale\n",
    "im_gray = cv2.cvtColor(contrast_img, cv2.COLOR_BGR2GRAY)\n",
    "#binarize the image using Otsu method to get the threshold, binarize to black and white(ie. 255)\n",
    "th, im_gray_th_otsu = cv2.threshold(im_gray, 128, 255, cv2.THRESH_OTSU)\n",
    "#resizing...\n",
    "resized = cv2.resize(im_gray_th_otsu, (32,32), interpolation = cv2.INTER_AREA)\n",
    "\n",
    "#print the threshold because I'm curious\n",
    "print(th)\n",
    "\n",
    "final_img = np.zeros(resized.shape, resized.dtype)\n",
    "for y in range(resized.shape[0]):\n",
    "    for x in range(resized.shape[1]):\n",
    "        if resized[y,x] < 125:     \n",
    "            final_img[y,x] = 255\n",
    "        else :\n",
    "            final_img[y,x] = 0\n",
    "\n",
    "#Show the images so we can compare them (use the cv2.imshow if you want to ee them all at once)\n",
    "#cv2.imshow('New Image', im_gray_th_otsu)\n",
    "#cv2.imshow('Original Image', image)\n",
    "#cv2.imshow('Contrast image', contrast_img)\n",
    "#cv2.imshow('resized Image', resized)\n",
    "#cv2.imshow('final Image', final_img)\n",
    "plt.imshow(image)\n",
    "plt.imshow(contrast_img)\n",
    "plt.imshow(im_gray_th_otsu, cmap='gray', vmin=0, vmax=255)\n",
    "plt.imshow(resized, cmap='gray', vmin=0, vmax=255)\n",
    "plt.imshow(final_img, cmap='gray', vmin=0, vmax=255)\n",
    "\n",
    "\n",
    "final_img = np.asarray(final_img).reshape([-1, 32, 32, 1])\n",
    "#final_img.reshape([1,32,32,1])\n",
    "#final_img = np.expand_dims(final_img, -1)\n",
    "\n",
    "predictions_single = recognizer.predict(final_img)\n",
    "print(predictions_single)\n",
    "\n",
    "#let's get our label\n",
    "label = 0\n",
    "for i in range(28):\n",
    "    if predictions_single[0][i] == 1:\n",
    "        label = i+1\n",
    "        break\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "Your letter is alif\n"
     ]
    }
   ],
   "source": [
    "print(label)\n",
    "outputMsg(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x24185630ac0>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(contrast_img)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x241856932b0>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(im_gray_th_otsu, cmap='gray', vmin=0, vmax=255)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x241856ea400>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(resized, cmap='gray', vmin=0, vmax=255)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['conv2d',\n",
       " 'conv2d_1',\n",
       " 'average_pooling2d',\n",
       " 'dropout',\n",
       " 'conv2d_2',\n",
       " 'conv2d_3',\n",
       " 'max_pooling2d',\n",
       " 'dropout_1',\n",
       " 'flatten',\n",
       " 'dense',\n",
       " 'dense_1',\n",
       " 'dropout_2',\n",
       " 'dense_2']"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer_names = [layer.name for layer in recognizer.layers]\n",
    "layer_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<tensorflow.python.keras.layers.convolutional.Conv2D at 0x241f6b75a90>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241f6a296d0>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x241f6883640>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x241f6a40430>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241f6bad070>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241f6a504c0>,\n",
       " <tensorflow.python.keras.layers.pooling.MaxPooling2D at 0x241f6a6c7f0>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x241f6aaea30>,\n",
       " <tensorflow.python.keras.layers.core.Flatten at 0x241f6a77880>,\n",
       " <tensorflow.python.keras.layers.core.Dense at 0x241f6a77dc0>,\n",
       " <tensorflow.python.keras.layers.core.Dense at 0x241f6a7fe80>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x241f6a97220>,\n",
       " <tensorflow.python.keras.layers.core.Dense at 0x241f6a14460>]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recognizer.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<tf.Tensor 'conv2d/Relu:0' shape=(?, 32, 32, 32) dtype=float32>,\n",
       " <tf.Tensor 'conv2d_1/Relu:0' shape=(?, 32, 32, 32) dtype=float32>,\n",
       " <tf.Tensor 'average_pooling2d/AvgPool:0' shape=(?, 16, 16, 32) dtype=float32>,\n",
       " <tf.Tensor 'dropout/cond/Merge:0' shape=(?, 16, 16, 32) dtype=float32>,\n",
       " <tf.Tensor 'conv2d_2/Relu:0' shape=(?, 16, 16, 64) dtype=float32>,\n",
       " <tf.Tensor 'conv2d_3/Relu:0' shape=(?, 16, 16, 64) dtype=float32>,\n",
       " <tf.Tensor 'max_pooling2d/MaxPool:0' shape=(?, 15, 15, 64) dtype=float32>,\n",
       " <tf.Tensor 'dropout_1/cond/Merge:0' shape=(?, 15, 15, 64) dtype=float32>,\n",
       " <tf.Tensor 'flatten/Reshape:0' shape=(?, 14400) dtype=float32>,\n",
       " <tf.Tensor 'dense/Relu:0' shape=(?, 256) dtype=float32>,\n",
       " <tf.Tensor 'dense_1/Relu:0' shape=(?, 256) dtype=float32>,\n",
       " <tf.Tensor 'dropout_2/cond/Merge:0' shape=(?, 256) dtype=float32>,\n",
       " <tf.Tensor 'dense_2/Softmax:0' shape=(?, 28) dtype=float32>]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer_outputs = [layer.output for layer in recognizer.layers]\n",
    "layer_outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_map_model = tf.keras.models.Model(inputs=recognizer.input, outputs=layer_outputs)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor 'conv2d_input:0' shape=(?, 32, 32, 1) dtype=float32>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recognizer.input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_maps = feature_map_model.predict(final_img)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 14400)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_maps[8].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.0024024 0.        0.        ... 0.        0.        0.       ]]\n"
     ]
    }
   ],
   "source": [
    "print(feature_maps[8])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see if it cares about image size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 32, 32, 1)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_img.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(364, 385)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "im_gray_th_otsu.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(im_gray_th_otsu)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "sizeTestingimg = np.asarray(im_gray_th_otsu).reshape([-1, 364, 385, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Error when checking input: expected conv2d_input to have shape (32, 32, 1) but got array with shape (364, 385, 1)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-59-27603ed65cac>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mfeature_maps\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfeature_map_model\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msizeTestingimg\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training_v1.py\u001b[0m in \u001b[0;36mpredict\u001b[1;34m(self, x, batch_size, verbose, steps, callbacks, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[0;32m    980\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    981\u001b[0m     \u001b[0mfunc\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_select_training_loop\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 982\u001b[1;33m     return func.predict(\n\u001b[0m\u001b[0;32m    983\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    984\u001b[0m         \u001b[0mx\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training_arrays_v1.py\u001b[0m in \u001b[0;36mpredict\u001b[1;34m(self, model, x, batch_size, verbose, steps, callbacks, **kwargs)\u001b[0m\n\u001b[0;32m    702\u001b[0m               **kwargs):\n\u001b[0;32m    703\u001b[0m     \u001b[0mbatch_size\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_validate_or_infer_batch_size\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msteps\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 704\u001b[1;33m     x, _, _ = model._standardize_user_data(\n\u001b[0m\u001b[0;32m    705\u001b[0m         x, check_steps=True, steps_name='steps', steps=steps)\n\u001b[0;32m    706\u001b[0m     return predict_loop(\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training_v1.py\u001b[0m in \u001b[0;36m_standardize_user_data\u001b[1;34m(self, x, y, sample_weight, class_weight, batch_size, check_steps, steps_name, steps, validation_split, shuffle, extract_tensors_from_dataset)\u001b[0m\n\u001b[0;32m   2328\u001b[0m       \u001b[1;32mreturn\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2329\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2330\u001b[1;33m     return self._standardize_tensors(\n\u001b[0m\u001b[0;32m   2331\u001b[0m         \u001b[0mx\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2332\u001b[0m         \u001b[0mrun_eagerly\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mrun_eagerly\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training_v1.py\u001b[0m in \u001b[0;36m_standardize_tensors\u001b[1;34m(self, x, y, sample_weight, run_eagerly, dict_inputs, is_dataset, class_weight, batch_size)\u001b[0m\n\u001b[0;32m   2356\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[0mdataset_ops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDatasetV1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdataset_ops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDatasetV2\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2357\u001b[0m       \u001b[1;31m# TODO(fchollet): run static checks with dataset output shape(s).\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2358\u001b[1;33m       x = training_utils_v1.standardize_input_data(\n\u001b[0m\u001b[0;32m   2359\u001b[0m           \u001b[0mx\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2360\u001b[0m           \u001b[0mfeed_input_names\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training_utils_v1.py\u001b[0m in \u001b[0;36mstandardize_input_data\u001b[1;34m(data, names, shapes, check_batch_axis, exception_prefix)\u001b[0m\n\u001b[0;32m    535\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0mdim\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mref_dim\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata_shape\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mshape\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    536\u001b[0m           \u001b[1;32mif\u001b[0m \u001b[0mref_dim\u001b[0m \u001b[1;33m!=\u001b[0m \u001b[0mdim\u001b[0m \u001b[1;32mand\u001b[0m \u001b[0mref_dim\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;32mNone\u001b[0m \u001b[1;32mand\u001b[0m \u001b[0mdim\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 537\u001b[1;33m             raise ValueError('Error when checking ' + exception_prefix +\n\u001b[0m\u001b[0;32m    538\u001b[0m                              \u001b[1;34m': expected '\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0mnames\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m+\u001b[0m \u001b[1;34m' to have shape '\u001b[0m \u001b[1;33m+\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    539\u001b[0m                              \u001b[0mstr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m+\u001b[0m \u001b[1;34m' but got array with shape '\u001b[0m \u001b[1;33m+\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: Error when checking input: expected conv2d_input to have shape (32, 32, 1) but got array with shape (364, 385, 1)"
     ]
    }
   ],
   "source": [
    "feature_maps = feature_map_model.predict(sizeTestingimg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yes it does... so we have to change the shapes of each segment to fit this shape..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.applications.inception_v3 import InceptionV3\n",
    "from tensorflow.keras.applications.inception_v3 import preprocess_input\n",
    "from tensorflow.keras.preprocessing import image\n",
    "from tensorflow.keras.preprocessing.image import img_to_array\n",
    "from sklearn.cluster import KMeans\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "import os\n",
    "import shutil"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/inception_v3/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
      "87916544/87910968 [==============================] - 108s 1us/step\n"
     ]
    }
   ],
   "source": [
    "testingmodel = InceptionV3(weights='imagenet', include_top=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<tensorflow.python.keras.engine.input_layer.InputLayer at 0x241856b3eb0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241856fbe80>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418c1fd1f0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418c1f0760>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418c1fd340>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418c203eb0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418c23a8e0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418c241d30>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418c2493d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418c272280>,\n",
       " <tensorflow.python.keras.layers.pooling.MaxPooling2D at 0x2418c2723d0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418c280130>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418c288340>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418c27afa0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418c2a4dc0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418c2ad9a0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418c2cf4f0>,\n",
       " <tensorflow.python.keras.layers.pooling.MaxPooling2D at 0x2418c2cf190>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e17df10>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e190b50>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e1ad9a0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e123850>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e1b3fd0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e1455b0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e1b90a0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e1511f0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e1e04f0>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x2418e2111c0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418c2cfe80>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e15ee20>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e1e0190>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e2110a0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e1159d0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e16d550>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e1f2430>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e22cac0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e126df0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e182460>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e2111f0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e21c250>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x2418e241a90>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e2d5730>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e2f15e0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e3055e0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e272580>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e305d60>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e298160>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e314310>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e2a23a0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e331a00>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x2418e362eb0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e256670>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e2a2340>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e3311f0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e3620a0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e24d3d0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e2bfee0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e359f70>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e37f8b0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e277fd0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e2cdf10>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e362160>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e393520>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x2418e3931f0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e426b80>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e42da90>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e457760>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e3c5400>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e4577c0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e3cba90>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e463130>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e3f5190>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e4852e0>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x2418e4b88b0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e3998e0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e3f5ee0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e485400>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e4b8310>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e3bed90>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e413550>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e4b0700>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e4d3eb0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e3c5370>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e420b80>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e4b80d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e4e60a0>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x2418e4e6940>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e51a370>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e53fca0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e54ae80>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e536b50>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e569c40>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e577fd0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e4ef430>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e57cc70>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e513fa0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e582670>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e51a2e0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e5abee0>,\n",
       " <tensorflow.python.keras.layers.pooling.MaxPooling2D at 0x2418e5ab8e0>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x2418e5abc40>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e66c2b0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e6e26d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e6f6f10>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e6fed30>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e70b1f0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e72a970>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e5e0040>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e730bb0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e5ea1f0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e73d880>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e60a310>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e760c40>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e610bb0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e759520>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e61c880>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e781490>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e640d00>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e78d1f0>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x2418e7996a0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e5cde50>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e63b520>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e78d160>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e7c2c70>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e5da9d0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e660130>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e7a8a90>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e7d13a0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e5da7c0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e66c1c0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e7be310>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e7e9d30>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x2418e7d4130>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e8b4d00>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e8c1760>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418c272be0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418c21ab20>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e211790>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241f3df3e20>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e820370>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e277700>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e8457f0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e370490>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e852160>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e6db5b0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e852c40>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e55fe50>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e86e880>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e5b4250>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e882f10>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418560a400>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x2418e95d430>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e7fc9d0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e87ae20>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418e82d220>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418d0f1820>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e819f40>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e89fdc0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418e85ed60>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24187033ac0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e820310>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418e8ad070>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418d0d2a30>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241855f8850>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24186027d30>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181b06eb0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181b15d90>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181b30c10>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181b36490>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181b3c670>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181b64b80>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24185598160>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181b64850>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181a9d5e0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181b74be0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181aaf5e0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181b8fa00>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181aafd60>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181b8f3a0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181abd100>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181bb4f70>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181ad8a00>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181bbf160>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x24181be8c40>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418ca13a60>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181ad81f0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181bbfeb0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181beec70>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418d0b1e80>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181afdf70>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181bdc5e0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181c0c5b0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418558f100>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181b06160>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181bee580>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181c206a0>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24181c198e0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181ce13a0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181cfb850>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181d121c0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181d12f10>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181d2e130>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181d3b670>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181c506d0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181d416d0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181c5c280>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181d48ac0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181c7bd00>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181d72eb0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181c813a0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181d72ca0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181c88ac0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181d7b9d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181cac160>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181d9e070>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x24182da0580>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181c26b50>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181cb1ca0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24181d9e430>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182da0130>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181c2c100>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181cbc460>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24181db2400>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182dbd6d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181c50250>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24181ce13d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182da02b0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182dce160>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24182dce070>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182e34310>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182e41130>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182e668e0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182e5ef70>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182e83dc0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182e8f850>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182dde520>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182e95a60>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182dfce20>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182e9d6d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182e04400>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182ec2520>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182e04490>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182ec23a0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182e27e80>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182ee8e80>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182e34610>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182ef2a00>,\n",
       " <tensorflow.python.keras.layers.pooling.MaxPooling2D at 0x24182ef2df0>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24182ef20d0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182fc4e50>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182fca760>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182fea400>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182f28e50>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182fe2850>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182f39790>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182fef520>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182f564f0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24183014760>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182f562b0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182f87580>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418301bcd0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24183045460>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x24183085700>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24182f06b80>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182f7ca60>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182f8d310>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418301ed60>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418306c9d0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418308c0a0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24182f1d550>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182f87130>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182fb1c10>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418304c8e0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241830771c0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x241830a28e0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24182f28cd0>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24182fb83d0>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24183077190>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241830b0e20>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x241830b0cd0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24183176be0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418318db20>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418319bc10>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241830daaf0>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241831a29d0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x241830edd90>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x241831bf460>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241831128e0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241831cb190>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418310a700>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x2418313f100>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241831d1b20>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241831fc430>,\n",
       " <tensorflow.python.keras.layers.pooling.AveragePooling2D at 0x24183244430>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x241830aa520>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24183131d60>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24183155a00>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x241831da9d0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x24183224d60>,\n",
       " <tensorflow.python.keras.layers.convolutional.Conv2D at 0x24183237ac0>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x241830e1d00>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418313f3d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24183170070>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241832049d0>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x2418322fa00>,\n",
       " <tensorflow.python.keras.layers.normalization.BatchNormalization at 0x2418325ad30>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x241830e1100>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24183169fa0>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x2418322f190>,\n",
       " <tensorflow.python.keras.layers.core.Activation at 0x24183268e20>,\n",
       " <tensorflow.python.keras.layers.merge.Concatenate at 0x24183268af0>]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testingmodel.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\rassa\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training.py:2325: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n",
      "  warnings.warn('`Model.state_updates` will be removed in a future version. '\n"
     ]
    }
   ],
   "source": [
    "img=image.load_img(r\"D:\\School\\MedTech\\Sophomore Pre-Eng\\ISS296\\CNN\\images\\2alif.jpg\",target_size=(224,224))\n",
    "x = img_to_array(img)\n",
    "x=np.expand_dims(x,axis=0)\n",
    "x=preprocess_input(x)\n",
    "testingmodeloutput = testingmodel.predict(x).flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(51200,)"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testingmodeloutput.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.        0.        0.        ... 0.3115046 1.2965667 0.       ]\n"
     ]
    }
   ],
   "source": [
    "print(testingmodeloutput)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Lets try k clustering with the current dataset of characters first"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                         | 0/3360 [00:00<?, ?it/s]C:\\Users\\rassa\\anaconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\training.py:2325: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n",
      "  warnings.warn('`Model.state_updates` will be removed in a future version. '\n",
      "100%|█████████████████████████████████████████████████████████████████████████████| 3360/3360 [00:17<00:00, 194.40it/s]\n"
     ]
    }
   ],
   "source": [
    "#trainPath = DIRECTORY + r\"\\Train Images 13440x32x32\\train\"\n",
    "#testPath  = DIRECTORY + r\"\\Test Images 3360x32x32\\test\"\n",
    "\n",
    "\n",
    "\n",
    "def image_feature(directory, sortingCriteria):\n",
    "    model = tf.keras.models.Model(inputs=recognizer.input, outputs=layer_outputs)\n",
    "    features = [];\n",
    "    img_name = [];\n",
    "    \n",
    "    imgFileNameList = os.listdir(directory)     #get list of directories in path (ie. list of image names)\n",
    "    imgFileNameList.sort(key = sortingCriteria) #sort list based on key\n",
    "    \n",
    "    for img in tqdm(imgFileNameList):  # iterate over the sorted list\n",
    "        #try:\n",
    "            #img_array = cv2.imread(os.path.join(directory, img) ,cv2.IMREAD_GRAYSCALE)  # convert image to array\n",
    "            #data.append([img_array])  # add this to our data\n",
    "        #except Exception as e:  # just in case there is a mistake.\n",
    "            #pass\n",
    "            \n",
    "    #for i in tqdm(direc):\n",
    "        #fname='cluster'+'/'+i\n",
    "        img_array = cv2.imread(os.path.join(directory, img) ,cv2.IMREAD_GRAYSCALE)   #img=image.load_img(fname,target_size=(224,224))\n",
    "        #x = img_to_array(img)\n",
    "        #x=np.expand_dims(x,axis=0)\n",
    "        #x=preprocess_input(x)\n",
    "        img_array=img_array.reshape([-1, 32, 32, 1])\n",
    "        LoFeatureMaps = model.predict(img_array)   #feat=model.predict(x)\n",
    "        #feat=feat.flatten()\n",
    "        resizedFlattenedFeatureMap = LoFeatureMaps[8].reshape([LoFeatureMaps[8].shape[1]])\n",
    "        features.append(resizedFlattenedFeatureMap) #features.append(feat)\n",
    "        \n",
    "        img_name.append(img)\n",
    "    return features,img_name\n",
    "\n",
    "\n",
    "LoTestImgFeatures,LoTestImgNames = image_feature(testPath, getImgId)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KMeans(n_clusters=28, random_state=40)"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Creating Clusters\n",
    "k = 28\n",
    "clusters = KMeans(k, random_state = 40)\n",
    "clusters.fit(LoTestImgFeatures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['id_1_label_1.png', 'id_2_label_1.png', 'id_3_label_2.png', 'id_4_label_2.png', 'id_5_label_3.png', 'id_6_label_3.png', 'id_7_label_4.png', 'id_8_label_4.png', 'id_9_label_5.png', 'id_10_label_5.png', 'id_11_label_6.png', 'id_12_label_6.png', 'id_13_label_7.png', 'id_14_label_7.png', 'id_15_label_8.png', 'id_16_label_8.png', 'id_17_label_9.png', 'id_18_label_9.png', 'id_19_label_10.png', 'id_20_label_10.png', 'id_21_label_11.png', 'id_22_label_11.png', 'id_23_label_12.png', 'id_24_label_12.png', 'id_25_label_13.png', 'id_26_label_13.png', 'id_27_label_14.png', 'id_28_label_14.png', 'id_29_label_15.png', 'id_30_label_15.png', 'id_31_label_16.png', 'id_32_label_16.png', 'id_33_label_17.png', 'id_34_label_17.png', 'id_35_label_18.png', 'id_36_label_18.png', 'id_37_label_19.png', 'id_38_label_19.png', 'id_39_label_20.png', 'id_40_label_20.png', 'id_41_label_21.png', 'id_42_label_21.png', 'id_43_label_22.png', 'id_44_label_22.png', 'id_45_label_23.png', 'id_46_label_23.png', 'id_47_label_24.png', 'id_48_label_24.png', 'id_49_label_25.png', 'id_50_label_25.png', 'id_51_label_26.png', 'id_52_label_26.png', 'id_53_label_27.png', 'id_54_label_27.png', 'id_55_label_28.png', 'id_56_label_28.png', 'id_57_label_1.png', 'id_58_label_1.png', 'id_59_label_2.png', 'id_60_label_2.png', 'id_61_label_3.png', 'id_62_label_3.png', 'id_63_label_4.png', 'id_64_label_4.png', 'id_65_label_5.png', 'id_66_label_5.png', 'id_67_label_6.png', 'id_68_label_6.png', 'id_69_label_7.png', 'id_70_label_7.png', 'id_71_label_8.png', 'id_72_label_8.png', 'id_73_label_9.png', 'id_74_label_9.png', 'id_75_label_10.png', 'id_76_label_10.png', 'id_77_label_11.png', 'id_78_label_11.png', 'id_79_label_12.png', 'id_80_label_12.png', 'id_81_label_13.png', 'id_82_label_13.png', 'id_83_label_14.png', 'id_84_label_14.png', 'id_85_label_15.png', 'id_86_label_15.png', 'id_87_label_16.png', 'id_88_label_16.png', 'id_89_label_17.png', 'id_90_label_17.png', 'id_91_label_18.png', 'id_92_label_18.png', 'id_93_label_19.png', 'id_94_label_19.png', 'id_95_label_20.png', 'id_96_label_20.png', 'id_97_label_21.png', 'id_98_label_21.png', 'id_99_label_22.png', 'id_100_label_22.png', 'id_101_label_23.png', 'id_102_label_23.png', 'id_103_label_24.png', 'id_104_label_24.png', 'id_105_label_25.png', 'id_106_label_25.png', 'id_107_label_26.png', 'id_108_label_26.png', 'id_109_label_27.png', 'id_110_label_27.png', 'id_111_label_28.png', 'id_112_label_28.png', 'id_113_label_1.png', 'id_114_label_1.png', 'id_115_label_2.png', 'id_116_label_2.png', 'id_117_label_3.png', 'id_118_label_3.png', 'id_119_label_4.png', 'id_120_label_4.png', 'id_121_label_5.png', 'id_122_label_5.png', 'id_123_label_6.png', 'id_124_label_6.png', 'id_125_label_7.png', 'id_126_label_7.png', 'id_127_label_8.png', 'id_128_label_8.png', 'id_129_label_9.png', 'id_130_label_9.png', 'id_131_label_10.png', 'id_132_label_10.png', 'id_133_label_11.png', 'id_134_label_11.png', 'id_135_label_12.png', 'id_136_label_12.png', 'id_137_label_13.png', 'id_138_label_13.png', 'id_139_label_14.png', 'id_140_label_14.png', 'id_141_label_15.png', 'id_142_label_15.png', 'id_143_label_16.png', 'id_144_label_16.png', 'id_145_label_17.png', 'id_146_label_17.png', 'id_147_label_18.png', 'id_148_label_18.png', 'id_149_label_19.png', 'id_150_label_19.png', 'id_151_label_20.png', 'id_152_label_20.png', 'id_153_label_21.png', 'id_154_label_21.png', 'id_155_label_22.png', 'id_156_label_22.png', 'id_157_label_23.png', 'id_158_label_23.png', 'id_159_label_24.png', 'id_160_label_24.png', 'id_161_label_25.png', 'id_162_label_25.png', 'id_163_label_26.png', 'id_164_label_26.png', 'id_165_label_27.png', 'id_166_label_27.png', 'id_167_label_28.png', 'id_168_label_28.png', 'id_169_label_1.png', 'id_170_label_1.png', 'id_171_label_2.png', 'id_172_label_2.png', 'id_173_label_3.png', 'id_174_label_3.png', 'id_175_label_4.png', 'id_176_label_4.png', 'id_177_label_5.png', 'id_178_label_5.png', 'id_179_label_6.png', 'id_180_label_6.png', 'id_181_label_7.png', 'id_182_label_7.png', 'id_183_label_8.png', 'id_184_label_8.png', 'id_185_label_9.png', 'id_186_label_9.png', 'id_187_label_10.png', 'id_188_label_10.png', 'id_189_label_11.png', 'id_190_label_11.png', 'id_191_label_12.png', 'id_192_label_12.png', 'id_193_label_13.png', 'id_194_label_13.png', 'id_195_label_14.png', 'id_196_label_14.png', 'id_197_label_15.png', 'id_198_label_15.png', 'id_199_label_16.png', 'id_200_label_16.png', 'id_201_label_17.png', 'id_202_label_17.png', 'id_203_label_18.png', 'id_204_label_18.png', 'id_205_label_19.png', 'id_206_label_19.png', 'id_207_label_20.png', 'id_208_label_20.png', 'id_209_label_21.png', 'id_210_label_21.png', 'id_211_label_22.png', 'id_212_label_22.png', 'id_213_label_23.png', 'id_214_label_23.png', 'id_215_label_24.png', 'id_216_label_24.png', 'id_217_label_25.png', 'id_218_label_25.png', 'id_219_label_26.png', 'id_220_label_26.png', 'id_221_label_27.png', 'id_222_label_27.png', 'id_223_label_28.png', 'id_224_label_28.png', 'id_225_label_1.png', 'id_226_label_1.png', 'id_227_label_2.png', 'id_228_label_2.png', 'id_229_label_3.png', 'id_230_label_3.png', 'id_231_label_4.png', 'id_232_label_4.png', 'id_233_label_5.png', 'id_234_label_5.png', 'id_235_label_6.png', 'id_236_label_6.png', 'id_237_label_7.png', 'id_238_label_7.png', 'id_239_label_8.png', 'id_240_label_8.png', 'id_241_label_9.png', 'id_242_label_9.png', 'id_243_label_10.png', 'id_244_label_10.png', 'id_245_label_11.png', 'id_246_label_11.png', 'id_247_label_12.png', 'id_248_label_12.png', 'id_249_label_13.png', 'id_250_label_13.png', 'id_251_label_14.png', 'id_252_label_14.png', 'id_253_label_15.png', 'id_254_label_15.png', 'id_255_label_16.png', 'id_256_label_16.png', 'id_257_label_17.png', 'id_258_label_17.png', 'id_259_label_18.png', 'id_260_label_18.png', 'id_261_label_19.png', 'id_262_label_19.png', 'id_263_label_20.png', 'id_264_label_20.png', 'id_265_label_21.png', 'id_266_label_21.png', 'id_267_label_22.png', 'id_268_label_22.png', 'id_269_label_23.png', 'id_270_label_23.png', 'id_271_label_24.png', 'id_272_label_24.png', 'id_273_label_25.png', 'id_274_label_25.png', 'id_275_label_26.png', 'id_276_label_26.png', 'id_277_label_27.png', 'id_278_label_27.png', 'id_279_label_28.png', 'id_280_label_28.png', 'id_281_label_1.png', 'id_282_label_1.png', 'id_283_label_2.png', 'id_284_label_2.png', 'id_285_label_3.png', 'id_286_label_3.png', 'id_287_label_4.png', 'id_288_label_4.png', 'id_289_label_5.png', 'id_290_label_5.png', 'id_291_label_6.png', 'id_292_label_6.png', 'id_293_label_7.png', 'id_294_label_7.png', 'id_295_label_8.png', 'id_296_label_8.png', 'id_297_label_9.png', 'id_298_label_9.png', 'id_299_label_10.png', 'id_300_label_10.png', 'id_301_label_11.png', 'id_302_label_11.png', 'id_303_label_12.png', 'id_304_label_12.png', 'id_305_label_13.png', 'id_306_label_13.png', 'id_307_label_14.png', 'id_308_label_14.png', 'id_309_label_15.png', 'id_310_label_15.png', 'id_311_label_16.png', 'id_312_label_16.png', 'id_313_label_17.png', 'id_314_label_17.png', 'id_315_label_18.png', 'id_316_label_18.png', 'id_317_label_19.png', 'id_318_label_19.png', 'id_319_label_20.png', 'id_320_label_20.png', 'id_321_label_21.png', 'id_322_label_21.png', 'id_323_label_22.png', 'id_324_label_22.png', 'id_325_label_23.png', 'id_326_label_23.png', 'id_327_label_24.png', 'id_328_label_24.png', 'id_329_label_25.png', 'id_330_label_25.png', 'id_331_label_26.png', 'id_332_label_26.png', 'id_333_label_27.png', 'id_334_label_27.png', 'id_335_label_28.png', 'id_336_label_28.png', 'id_337_label_1.png', 'id_338_label_1.png', 'id_339_label_2.png', 'id_340_label_2.png', 'id_341_label_3.png', 'id_342_label_3.png', 'id_343_label_4.png', 'id_344_label_4.png', 'id_345_label_5.png', 'id_346_label_5.png', 'id_347_label_6.png', 'id_348_label_6.png', 'id_349_label_7.png', 'id_350_label_7.png', 'id_351_label_8.png', 'id_352_label_8.png', 'id_353_label_9.png', 'id_354_label_9.png', 'id_355_label_10.png', 'id_356_label_10.png', 'id_357_label_11.png', 'id_358_label_11.png', 'id_359_label_12.png', 'id_360_label_12.png', 'id_361_label_13.png', 'id_362_label_13.png', 'id_363_label_14.png', 'id_364_label_14.png', 'id_365_label_15.png', 'id_366_label_15.png', 'id_367_label_16.png', 'id_368_label_16.png', 'id_369_label_17.png', 'id_370_label_17.png', 'id_371_label_18.png', 'id_372_label_18.png', 'id_373_label_19.png', 'id_374_label_19.png', 'id_375_label_20.png', 'id_376_label_20.png', 'id_377_label_21.png', 'id_378_label_21.png', 'id_379_label_22.png', 'id_380_label_22.png', 'id_381_label_23.png', 'id_382_label_23.png', 'id_383_label_24.png', 'id_384_label_24.png', 'id_385_label_25.png', 'id_386_label_25.png', 'id_387_label_26.png', 'id_388_label_26.png', 'id_389_label_27.png', 'id_390_label_27.png', 'id_391_label_28.png', 'id_392_label_28.png', 'id_393_label_1.png', 'id_394_label_1.png', 'id_395_label_2.png', 'id_396_label_2.png', 'id_397_label_3.png', 'id_398_label_3.png', 'id_399_label_4.png', 'id_400_label_4.png', 'id_401_label_5.png', 'id_402_label_5.png', 'id_403_label_6.png', 'id_404_label_6.png', 'id_405_label_7.png', 'id_406_label_7.png', 'id_407_label_8.png', 'id_408_label_8.png', 'id_409_label_9.png', 'id_410_label_9.png', 'id_411_label_10.png', 'id_412_label_10.png', 'id_413_label_11.png', 'id_414_label_11.png', 'id_415_label_12.png', 'id_416_label_12.png', 'id_417_label_13.png', 'id_418_label_13.png', 'id_419_label_14.png', 'id_420_label_14.png', 'id_421_label_15.png', 'id_422_label_15.png', 'id_423_label_16.png', 'id_424_label_16.png', 'id_425_label_17.png', 'id_426_label_17.png', 'id_427_label_18.png', 'id_428_label_18.png', 'id_429_label_19.png', 'id_430_label_19.png', 'id_431_label_20.png', 'id_432_label_20.png', 'id_433_label_21.png', 'id_434_label_21.png', 'id_435_label_22.png', 'id_436_label_22.png', 'id_437_label_23.png', 'id_438_label_23.png', 'id_439_label_24.png', 'id_440_label_24.png', 'id_441_label_25.png', 'id_442_label_25.png', 'id_443_label_26.png', 'id_444_label_26.png', 'id_445_label_27.png', 'id_446_label_27.png', 'id_447_label_28.png', 'id_448_label_28.png', 'id_449_label_1.png', 'id_450_label_1.png', 'id_451_label_2.png', 'id_452_label_2.png', 'id_453_label_3.png', 'id_454_label_3.png', 'id_455_label_4.png', 'id_456_label_4.png', 'id_457_label_5.png', 'id_458_label_5.png', 'id_459_label_6.png', 'id_460_label_6.png', 'id_461_label_7.png', 'id_462_label_7.png', 'id_463_label_8.png', 'id_464_label_8.png', 'id_465_label_9.png', 'id_466_label_9.png', 'id_467_label_10.png', 'id_468_label_10.png', 'id_469_label_11.png', 'id_470_label_11.png', 'id_471_label_12.png', 'id_472_label_12.png', 'id_473_label_13.png', 'id_474_label_13.png', 'id_475_label_14.png', 'id_476_label_14.png', 'id_477_label_15.png', 'id_478_label_15.png', 'id_479_label_16.png', 'id_480_label_16.png', 'id_481_label_17.png', 'id_482_label_17.png', 'id_483_label_18.png', 'id_484_label_18.png', 'id_485_label_19.png', 'id_486_label_19.png', 'id_487_label_20.png', 'id_488_label_20.png', 'id_489_label_21.png', 'id_490_label_21.png', 'id_491_label_22.png', 'id_492_label_22.png', 'id_493_label_23.png', 'id_494_label_23.png', 'id_495_label_24.png', 'id_496_label_24.png', 'id_497_label_25.png', 'id_498_label_25.png', 'id_499_label_26.png', 'id_500_label_26.png', 'id_501_label_27.png', 'id_502_label_27.png', 'id_503_label_28.png', 'id_504_label_28.png', 'id_505_label_1.png', 'id_506_label_1.png', 'id_507_label_2.png', 'id_508_label_2.png', 'id_509_label_3.png', 'id_510_label_3.png', 'id_511_label_4.png', 'id_512_label_4.png', 'id_513_label_5.png', 'id_514_label_5.png', 'id_515_label_6.png', 'id_516_label_6.png', 'id_517_label_7.png', 'id_518_label_7.png', 'id_519_label_8.png', 'id_520_label_8.png', 'id_521_label_9.png', 'id_522_label_9.png', 'id_523_label_10.png', 'id_524_label_10.png', 'id_525_label_11.png', 'id_526_label_11.png', 'id_527_label_12.png', 'id_528_label_12.png', 'id_529_label_13.png', 'id_530_label_13.png', 'id_531_label_14.png', 'id_532_label_14.png', 'id_533_label_15.png', 'id_534_label_15.png', 'id_535_label_16.png', 'id_536_label_16.png', 'id_537_label_17.png', 'id_538_label_17.png', 'id_539_label_18.png', 'id_540_label_18.png', 'id_541_label_19.png', 'id_542_label_19.png', 'id_543_label_20.png', 'id_544_label_20.png', 'id_545_label_21.png', 'id_546_label_21.png', 'id_547_label_22.png', 'id_548_label_22.png', 'id_549_label_23.png', 'id_550_label_23.png', 'id_551_label_24.png', 'id_552_label_24.png', 'id_553_label_25.png', 'id_554_label_25.png', 'id_555_label_26.png', 'id_556_label_26.png', 'id_557_label_27.png', 'id_558_label_27.png', 'id_559_label_28.png', 'id_560_label_28.png', 'id_561_label_1.png', 'id_562_label_1.png', 'id_563_label_2.png', 'id_564_label_2.png', 'id_565_label_3.png', 'id_566_label_3.png', 'id_567_label_4.png', 'id_568_label_4.png', 'id_569_label_5.png', 'id_570_label_5.png', 'id_571_label_6.png', 'id_572_label_6.png', 'id_573_label_7.png', 'id_574_label_7.png', 'id_575_label_8.png', 'id_576_label_8.png', 'id_577_label_9.png', 'id_578_label_9.png', 'id_579_label_10.png', 'id_580_label_10.png', 'id_581_label_11.png', 'id_582_label_11.png', 'id_583_label_12.png', 'id_584_label_12.png', 'id_585_label_13.png', 'id_586_label_13.png', 'id_587_label_14.png', 'id_588_label_14.png', 'id_589_label_15.png', 'id_590_label_15.png', 'id_591_label_16.png', 'id_592_label_16.png', 'id_593_label_17.png', 'id_594_label_17.png', 'id_595_label_18.png', 'id_596_label_18.png', 'id_597_label_19.png', 'id_598_label_19.png', 'id_599_label_20.png', 'id_600_label_20.png', 'id_601_label_21.png', 'id_602_label_21.png', 'id_603_label_22.png', 'id_604_label_22.png', 'id_605_label_23.png', 'id_606_label_23.png', 'id_607_label_24.png', 'id_608_label_24.png', 'id_609_label_25.png', 'id_610_label_25.png', 'id_611_label_26.png', 'id_612_label_26.png', 'id_613_label_27.png', 'id_614_label_27.png', 'id_615_label_28.png', 'id_616_label_28.png', 'id_617_label_1.png', 'id_618_label_1.png', 'id_619_label_2.png', 'id_620_label_2.png', 'id_621_label_3.png', 'id_622_label_3.png', 'id_623_label_4.png', 'id_624_label_4.png', 'id_625_label_5.png', 'id_626_label_5.png', 'id_627_label_6.png', 'id_628_label_6.png', 'id_629_label_7.png', 'id_630_label_7.png', 'id_631_label_8.png', 'id_632_label_8.png', 'id_633_label_9.png', 'id_634_label_9.png', 'id_635_label_10.png', 'id_636_label_10.png', 'id_637_label_11.png', 'id_638_label_11.png', 'id_639_label_12.png', 'id_640_label_12.png', 'id_641_label_13.png', 'id_642_label_13.png', 'id_643_label_14.png', 'id_644_label_14.png', 'id_645_label_15.png', 'id_646_label_15.png', 'id_647_label_16.png', 'id_648_label_16.png', 'id_649_label_17.png', 'id_650_label_17.png', 'id_651_label_18.png', 'id_652_label_18.png', 'id_653_label_19.png', 'id_654_label_19.png', 'id_655_label_20.png', 'id_656_label_20.png', 'id_657_label_21.png', 'id_658_label_21.png', 'id_659_label_22.png', 'id_660_label_22.png', 'id_661_label_23.png', 'id_662_label_23.png', 'id_663_label_24.png', 'id_664_label_24.png', 'id_665_label_25.png', 'id_666_label_25.png', 'id_667_label_26.png', 'id_668_label_26.png', 'id_669_label_27.png', 'id_670_label_27.png', 'id_671_label_28.png', 'id_672_label_28.png', 'id_673_label_1.png', 'id_674_label_1.png', 'id_675_label_2.png', 'id_676_label_2.png', 'id_677_label_3.png', 'id_678_label_3.png', 'id_679_label_4.png', 'id_680_label_4.png', 'id_681_label_5.png', 'id_682_label_5.png', 'id_683_label_6.png', 'id_684_label_6.png', 'id_685_label_7.png', 'id_686_label_7.png', 'id_687_label_8.png', 'id_688_label_8.png', 'id_689_label_9.png', 'id_690_label_9.png', 'id_691_label_10.png', 'id_692_label_10.png', 'id_693_label_11.png', 'id_694_label_11.png', 'id_695_label_12.png', 'id_696_label_12.png', 'id_697_label_13.png', 'id_698_label_13.png', 'id_699_label_14.png', 'id_700_label_14.png', 'id_701_label_15.png', 'id_702_label_15.png', 'id_703_label_16.png', 'id_704_label_16.png', 'id_705_label_17.png', 'id_706_label_17.png', 'id_707_label_18.png', 'id_708_label_18.png', 'id_709_label_19.png', 'id_710_label_19.png', 'id_711_label_20.png', 'id_712_label_20.png', 'id_713_label_21.png', 'id_714_label_21.png', 'id_715_label_22.png', 'id_716_label_22.png', 'id_717_label_23.png', 'id_718_label_23.png', 'id_719_label_24.png', 'id_720_label_24.png', 'id_721_label_25.png', 'id_722_label_25.png', 'id_723_label_26.png', 'id_724_label_26.png', 'id_725_label_27.png', 'id_726_label_27.png', 'id_727_label_28.png', 'id_728_label_28.png', 'id_729_label_1.png', 'id_730_label_1.png', 'id_731_label_2.png', 'id_732_label_2.png', 'id_733_label_3.png', 'id_734_label_3.png', 'id_735_label_4.png', 'id_736_label_4.png', 'id_737_label_5.png', 'id_738_label_5.png', 'id_739_label_6.png', 'id_740_label_6.png', 'id_741_label_7.png', 'id_742_label_7.png', 'id_743_label_8.png', 'id_744_label_8.png', 'id_745_label_9.png', 'id_746_label_9.png', 'id_747_label_10.png', 'id_748_label_10.png', 'id_749_label_11.png', 'id_750_label_11.png', 'id_751_label_12.png', 'id_752_label_12.png', 'id_753_label_13.png', 'id_754_label_13.png', 'id_755_label_14.png', 'id_756_label_14.png', 'id_757_label_15.png', 'id_758_label_15.png', 'id_759_label_16.png', 'id_760_label_16.png', 'id_761_label_17.png', 'id_762_label_17.png', 'id_763_label_18.png', 'id_764_label_18.png', 'id_765_label_19.png', 'id_766_label_19.png', 'id_767_label_20.png', 'id_768_label_20.png', 'id_769_label_21.png', 'id_770_label_21.png', 'id_771_label_22.png', 'id_772_label_22.png', 'id_773_label_23.png', 'id_774_label_23.png', 'id_775_label_24.png', 'id_776_label_24.png', 'id_777_label_25.png', 'id_778_label_25.png', 'id_779_label_26.png', 'id_780_label_26.png', 'id_781_label_27.png', 'id_782_label_27.png', 'id_783_label_28.png', 'id_784_label_28.png', 'id_785_label_1.png', 'id_786_label_1.png', 'id_787_label_2.png', 'id_788_label_2.png', 'id_789_label_3.png', 'id_790_label_3.png', 'id_791_label_4.png', 'id_792_label_4.png', 'id_793_label_5.png', 'id_794_label_5.png', 'id_795_label_6.png', 'id_796_label_6.png', 'id_797_label_7.png', 'id_798_label_7.png', 'id_799_label_8.png', 'id_800_label_8.png', 'id_801_label_9.png', 'id_802_label_9.png', 'id_803_label_10.png', 'id_804_label_10.png', 'id_805_label_11.png', 'id_806_label_11.png', 'id_807_label_12.png', 'id_808_label_12.png', 'id_809_label_13.png', 'id_810_label_13.png', 'id_811_label_14.png', 'id_812_label_14.png', 'id_813_label_15.png', 'id_814_label_15.png', 'id_815_label_16.png', 'id_816_label_16.png', 'id_817_label_17.png', 'id_818_label_17.png', 'id_819_label_18.png', 'id_820_label_18.png', 'id_821_label_19.png', 'id_822_label_19.png', 'id_823_label_20.png', 'id_824_label_20.png', 'id_825_label_21.png', 'id_826_label_21.png', 'id_827_label_22.png', 'id_828_label_22.png', 'id_829_label_23.png', 'id_830_label_23.png', 'id_831_label_24.png', 'id_832_label_24.png', 'id_833_label_25.png', 'id_834_label_25.png', 'id_835_label_26.png', 'id_836_label_26.png', 'id_837_label_27.png', 'id_838_label_27.png', 'id_839_label_28.png', 'id_840_label_28.png', 'id_841_label_1.png', 'id_842_label_1.png', 'id_843_label_2.png', 'id_844_label_2.png', 'id_845_label_3.png', 'id_846_label_3.png', 'id_847_label_4.png', 'id_848_label_4.png', 'id_849_label_5.png', 'id_850_label_5.png', 'id_851_label_6.png', 'id_852_label_6.png', 'id_853_label_7.png', 'id_854_label_7.png', 'id_855_label_8.png', 'id_856_label_8.png', 'id_857_label_9.png', 'id_858_label_9.png', 'id_859_label_10.png', 'id_860_label_10.png', 'id_861_label_11.png', 'id_862_label_11.png', 'id_863_label_12.png', 'id_864_label_12.png', 'id_865_label_13.png', 'id_866_label_13.png', 'id_867_label_14.png', 'id_868_label_14.png', 'id_869_label_15.png', 'id_870_label_15.png', 'id_871_label_16.png', 'id_872_label_16.png', 'id_873_label_17.png', 'id_874_label_17.png', 'id_875_label_18.png', 'id_876_label_18.png', 'id_877_label_19.png', 'id_878_label_19.png', 'id_879_label_20.png', 'id_880_label_20.png', 'id_881_label_21.png', 'id_882_label_21.png', 'id_883_label_22.png', 'id_884_label_22.png', 'id_885_label_23.png', 'id_886_label_23.png', 'id_887_label_24.png', 'id_888_label_24.png', 'id_889_label_25.png', 'id_890_label_25.png', 'id_891_label_26.png', 'id_892_label_26.png', 'id_893_label_27.png', 'id_894_label_27.png', 'id_895_label_28.png', 'id_896_label_28.png', 'id_897_label_1.png', 'id_898_label_1.png', 'id_899_label_2.png', 'id_900_label_2.png', 'id_901_label_3.png', 'id_902_label_3.png', 'id_903_label_4.png', 'id_904_label_4.png', 'id_905_label_5.png', 'id_906_label_5.png', 'id_907_label_6.png', 'id_908_label_6.png', 'id_909_label_7.png', 'id_910_label_7.png', 'id_911_label_8.png', 'id_912_label_8.png', 'id_913_label_9.png', 'id_914_label_9.png', 'id_915_label_10.png', 'id_916_label_10.png', 'id_917_label_11.png', 'id_918_label_11.png', 'id_919_label_12.png', 'id_920_label_12.png', 'id_921_label_13.png', 'id_922_label_13.png', 'id_923_label_14.png', 'id_924_label_14.png', 'id_925_label_15.png', 'id_926_label_15.png', 'id_927_label_16.png', 'id_928_label_16.png', 'id_929_label_17.png', 'id_930_label_17.png', 'id_931_label_18.png', 'id_932_label_18.png', 'id_933_label_19.png', 'id_934_label_19.png', 'id_935_label_20.png', 'id_936_label_20.png', 'id_937_label_21.png', 'id_938_label_21.png', 'id_939_label_22.png', 'id_940_label_22.png', 'id_941_label_23.png', 'id_942_label_23.png', 'id_943_label_24.png', 'id_944_label_24.png', 'id_945_label_25.png', 'id_946_label_25.png', 'id_947_label_26.png', 'id_948_label_26.png', 'id_949_label_27.png', 'id_950_label_27.png', 'id_951_label_28.png', 'id_952_label_28.png', 'id_953_label_1.png', 'id_954_label_1.png', 'id_955_label_2.png', 'id_956_label_2.png', 'id_957_label_3.png', 'id_958_label_3.png', 'id_959_label_4.png', 'id_960_label_4.png', 'id_961_label_5.png', 'id_962_label_5.png', 'id_963_label_6.png', 'id_964_label_6.png', 'id_965_label_7.png', 'id_966_label_7.png', 'id_967_label_8.png', 'id_968_label_8.png', 'id_969_label_9.png', 'id_970_label_9.png', 'id_971_label_10.png', 'id_972_label_10.png', 'id_973_label_11.png', 'id_974_label_11.png', 'id_975_label_12.png', 'id_976_label_12.png', 'id_977_label_13.png', 'id_978_label_13.png', 'id_979_label_14.png', 'id_980_label_14.png', 'id_981_label_15.png', 'id_982_label_15.png', 'id_983_label_16.png', 'id_984_label_16.png', 'id_985_label_17.png', 'id_986_label_17.png', 'id_987_label_18.png', 'id_988_label_18.png', 'id_989_label_19.png', 'id_990_label_19.png', 'id_991_label_20.png', 'id_992_label_20.png', 'id_993_label_21.png', 'id_994_label_21.png', 'id_995_label_22.png', 'id_996_label_22.png', 'id_997_label_23.png', 'id_998_label_23.png', 'id_999_label_24.png', 'id_1000_label_24.png', 'id_1001_label_25.png', 'id_1002_label_25.png', 'id_1003_label_26.png', 'id_1004_label_26.png', 'id_1005_label_27.png', 'id_1006_label_27.png', 'id_1007_label_28.png', 'id_1008_label_28.png', 'id_1009_label_1.png', 'id_1010_label_1.png', 'id_1011_label_2.png', 'id_1012_label_2.png', 'id_1013_label_3.png', 'id_1014_label_3.png', 'id_1015_label_4.png', 'id_1016_label_4.png', 'id_1017_label_5.png', 'id_1018_label_5.png', 'id_1019_label_6.png', 'id_1020_label_6.png', 'id_1021_label_7.png', 'id_1022_label_7.png', 'id_1023_label_8.png', 'id_1024_label_8.png', 'id_1025_label_9.png', 'id_1026_label_9.png', 'id_1027_label_10.png', 'id_1028_label_10.png', 'id_1029_label_11.png', 'id_1030_label_11.png', 'id_1031_label_12.png', 'id_1032_label_12.png', 'id_1033_label_13.png', 'id_1034_label_13.png', 'id_1035_label_14.png', 'id_1036_label_14.png', 'id_1037_label_15.png', 'id_1038_label_15.png', 'id_1039_label_16.png', 'id_1040_label_16.png', 'id_1041_label_17.png', 'id_1042_label_17.png', 'id_1043_label_18.png', 'id_1044_label_18.png', 'id_1045_label_19.png', 'id_1046_label_19.png', 'id_1047_label_20.png', 'id_1048_label_20.png', 'id_1049_label_21.png', 'id_1050_label_21.png', 'id_1051_label_22.png', 'id_1052_label_22.png', 'id_1053_label_23.png', 'id_1054_label_23.png', 'id_1055_label_24.png', 'id_1056_label_24.png', 'id_1057_label_25.png', 'id_1058_label_25.png', 'id_1059_label_26.png', 'id_1060_label_26.png', 'id_1061_label_27.png', 'id_1062_label_27.png', 'id_1063_label_28.png', 'id_1064_label_28.png', 'id_1065_label_1.png', 'id_1066_label_1.png', 'id_1067_label_2.png', 'id_1068_label_2.png', 'id_1069_label_3.png', 'id_1070_label_3.png', 'id_1071_label_4.png', 'id_1072_label_4.png', 'id_1073_label_5.png', 'id_1074_label_5.png', 'id_1075_label_6.png', 'id_1076_label_6.png', 'id_1077_label_7.png', 'id_1078_label_7.png', 'id_1079_label_8.png', 'id_1080_label_8.png', 'id_1081_label_9.png', 'id_1082_label_9.png', 'id_1083_label_10.png', 'id_1084_label_10.png', 'id_1085_label_11.png', 'id_1086_label_11.png', 'id_1087_label_12.png', 'id_1088_label_12.png', 'id_1089_label_13.png', 'id_1090_label_13.png', 'id_1091_label_14.png', 'id_1092_label_14.png', 'id_1093_label_15.png', 'id_1094_label_15.png', 'id_1095_label_16.png', 'id_1096_label_16.png', 'id_1097_label_17.png', 'id_1098_label_17.png', 'id_1099_label_18.png', 'id_1100_label_18.png', 'id_1101_label_19.png', 'id_1102_label_19.png', 'id_1103_label_20.png', 'id_1104_label_20.png', 'id_1105_label_21.png', 'id_1106_label_21.png', 'id_1107_label_22.png', 'id_1108_label_22.png', 'id_1109_label_23.png', 'id_1110_label_23.png', 'id_1111_label_24.png', 'id_1112_label_24.png', 'id_1113_label_25.png', 'id_1114_label_25.png', 'id_1115_label_26.png', 'id_1116_label_26.png', 'id_1117_label_27.png', 'id_1118_label_27.png', 'id_1119_label_28.png', 'id_1120_label_28.png', 'id_1121_label_1.png', 'id_1122_label_1.png', 'id_1123_label_2.png', 'id_1124_label_2.png', 'id_1125_label_3.png', 'id_1126_label_3.png', 'id_1127_label_4.png', 'id_1128_label_4.png', 'id_1129_label_5.png', 'id_1130_label_5.png', 'id_1131_label_6.png', 'id_1132_label_6.png', 'id_1133_label_7.png', 'id_1134_label_7.png', 'id_1135_label_8.png', 'id_1136_label_8.png', 'id_1137_label_9.png', 'id_1138_label_9.png', 'id_1139_label_10.png', 'id_1140_label_10.png', 'id_1141_label_11.png', 'id_1142_label_11.png', 'id_1143_label_12.png', 'id_1144_label_12.png', 'id_1145_label_13.png', 'id_1146_label_13.png', 'id_1147_label_14.png', 'id_1148_label_14.png', 'id_1149_label_15.png', 'id_1150_label_15.png', 'id_1151_label_16.png', 'id_1152_label_16.png', 'id_1153_label_17.png', 'id_1154_label_17.png', 'id_1155_label_18.png', 'id_1156_label_18.png', 'id_1157_label_19.png', 'id_1158_label_19.png', 'id_1159_label_20.png', 'id_1160_label_20.png', 'id_1161_label_21.png', 'id_1162_label_21.png', 'id_1163_label_22.png', 'id_1164_label_22.png', 'id_1165_label_23.png', 'id_1166_label_23.png', 'id_1167_label_24.png', 'id_1168_label_24.png', 'id_1169_label_25.png', 'id_1170_label_25.png', 'id_1171_label_26.png', 'id_1172_label_26.png', 'id_1173_label_27.png', 'id_1174_label_27.png', 'id_1175_label_28.png', 'id_1176_label_28.png', 'id_1177_label_1.png', 'id_1178_label_1.png', 'id_1179_label_2.png', 'id_1180_label_2.png', 'id_1181_label_3.png', 'id_1182_label_3.png', 'id_1183_label_4.png', 'id_1184_label_4.png', 'id_1185_label_5.png', 'id_1186_label_5.png', 'id_1187_label_6.png', 'id_1188_label_6.png', 'id_1189_label_7.png', 'id_1190_label_7.png', 'id_1191_label_8.png', 'id_1192_label_8.png', 'id_1193_label_9.png', 'id_1194_label_9.png', 'id_1195_label_10.png', 'id_1196_label_10.png', 'id_1197_label_11.png', 'id_1198_label_11.png', 'id_1199_label_12.png', 'id_1200_label_12.png', 'id_1201_label_13.png', 'id_1202_label_13.png', 'id_1203_label_14.png', 'id_1204_label_14.png', 'id_1205_label_15.png', 'id_1206_label_15.png', 'id_1207_label_16.png', 'id_1208_label_16.png', 'id_1209_label_17.png', 'id_1210_label_17.png', 'id_1211_label_18.png', 'id_1212_label_18.png', 'id_1213_label_19.png', 'id_1214_label_19.png', 'id_1215_label_20.png', 'id_1216_label_20.png', 'id_1217_label_21.png', 'id_1218_label_21.png', 'id_1219_label_22.png', 'id_1220_label_22.png', 'id_1221_label_23.png', 'id_1222_label_23.png', 'id_1223_label_24.png', 'id_1224_label_24.png', 'id_1225_label_25.png', 'id_1226_label_25.png', 'id_1227_label_26.png', 'id_1228_label_26.png', 'id_1229_label_27.png', 'id_1230_label_27.png', 'id_1231_label_28.png', 'id_1232_label_28.png', 'id_1233_label_1.png', 'id_1234_label_1.png', 'id_1235_label_2.png', 'id_1236_label_2.png', 'id_1237_label_3.png', 'id_1238_label_3.png', 'id_1239_label_4.png', 'id_1240_label_4.png', 'id_1241_label_5.png', 'id_1242_label_5.png', 'id_1243_label_6.png', 'id_1244_label_6.png', 'id_1245_label_7.png', 'id_1246_label_7.png', 'id_1247_label_8.png', 'id_1248_label_8.png', 'id_1249_label_9.png', 'id_1250_label_9.png', 'id_1251_label_10.png', 'id_1252_label_10.png', 'id_1253_label_11.png', 'id_1254_label_11.png', 'id_1255_label_12.png', 'id_1256_label_12.png', 'id_1257_label_13.png', 'id_1258_label_13.png', 'id_1259_label_14.png', 'id_1260_label_14.png', 'id_1261_label_15.png', 'id_1262_label_15.png', 'id_1263_label_16.png', 'id_1264_label_16.png', 'id_1265_label_17.png', 'id_1266_label_17.png', 'id_1267_label_18.png', 'id_1268_label_18.png', 'id_1269_label_19.png', 'id_1270_label_19.png', 'id_1271_label_20.png', 'id_1272_label_20.png', 'id_1273_label_21.png', 'id_1274_label_21.png', 'id_1275_label_22.png', 'id_1276_label_22.png', 'id_1277_label_23.png', 'id_1278_label_23.png', 'id_1279_label_24.png', 'id_1280_label_24.png', 'id_1281_label_25.png', 'id_1282_label_25.png', 'id_1283_label_26.png', 'id_1284_label_26.png', 'id_1285_label_27.png', 'id_1286_label_27.png', 'id_1287_label_28.png', 'id_1288_label_28.png', 'id_1289_label_1.png', 'id_1290_label_1.png', 'id_1291_label_2.png', 'id_1292_label_2.png', 'id_1293_label_3.png', 'id_1294_label_3.png', 'id_1295_label_4.png', 'id_1296_label_4.png', 'id_1297_label_5.png', 'id_1298_label_5.png', 'id_1299_label_6.png', 'id_1300_label_6.png', 'id_1301_label_7.png', 'id_1302_label_7.png', 'id_1303_label_8.png', 'id_1304_label_8.png', 'id_1305_label_9.png', 'id_1306_label_9.png', 'id_1307_label_10.png', 'id_1308_label_10.png', 'id_1309_label_11.png', 'id_1310_label_11.png', 'id_1311_label_12.png', 'id_1312_label_12.png', 'id_1313_label_13.png', 'id_1314_label_13.png', 'id_1315_label_14.png', 'id_1316_label_14.png', 'id_1317_label_15.png', 'id_1318_label_15.png', 'id_1319_label_16.png', 'id_1320_label_16.png', 'id_1321_label_17.png', 'id_1322_label_17.png', 'id_1323_label_18.png', 'id_1324_label_18.png', 'id_1325_label_19.png', 'id_1326_label_19.png', 'id_1327_label_20.png', 'id_1328_label_20.png', 'id_1329_label_21.png', 'id_1330_label_21.png', 'id_1331_label_22.png', 'id_1332_label_22.png', 'id_1333_label_23.png', 'id_1334_label_23.png', 'id_1335_label_24.png', 'id_1336_label_24.png', 'id_1337_label_25.png', 'id_1338_label_25.png', 'id_1339_label_26.png', 'id_1340_label_26.png', 'id_1341_label_27.png', 'id_1342_label_27.png', 'id_1343_label_28.png', 'id_1344_label_28.png', 'id_1345_label_1.png', 'id_1346_label_1.png', 'id_1347_label_2.png', 'id_1348_label_2.png', 'id_1349_label_3.png', 'id_1350_label_3.png', 'id_1351_label_4.png', 'id_1352_label_4.png', 'id_1353_label_5.png', 'id_1354_label_5.png', 'id_1355_label_6.png', 'id_1356_label_6.png', 'id_1357_label_7.png', 'id_1358_label_7.png', 'id_1359_label_8.png', 'id_1360_label_8.png', 'id_1361_label_9.png', 'id_1362_label_9.png', 'id_1363_label_10.png', 'id_1364_label_10.png', 'id_1365_label_11.png', 'id_1366_label_11.png', 'id_1367_label_12.png', 'id_1368_label_12.png', 'id_1369_label_13.png', 'id_1370_label_13.png', 'id_1371_label_14.png', 'id_1372_label_14.png', 'id_1373_label_15.png', 'id_1374_label_15.png', 'id_1375_label_16.png', 'id_1376_label_16.png', 'id_1377_label_17.png', 'id_1378_label_17.png', 'id_1379_label_18.png', 'id_1380_label_18.png', 'id_1381_label_19.png', 'id_1382_label_19.png', 'id_1383_label_20.png', 'id_1384_label_20.png', 'id_1385_label_21.png', 'id_1386_label_21.png', 'id_1387_label_22.png', 'id_1388_label_22.png', 'id_1389_label_23.png', 'id_1390_label_23.png', 'id_1391_label_24.png', 'id_1392_label_24.png', 'id_1393_label_25.png', 'id_1394_label_25.png', 'id_1395_label_26.png', 'id_1396_label_26.png', 'id_1397_label_27.png', 'id_1398_label_27.png', 'id_1399_label_28.png', 'id_1400_label_28.png', 'id_1401_label_1.png', 'id_1402_label_1.png', 'id_1403_label_2.png', 'id_1404_label_2.png', 'id_1405_label_3.png', 'id_1406_label_3.png', 'id_1407_label_4.png', 'id_1408_label_4.png', 'id_1409_label_5.png', 'id_1410_label_5.png', 'id_1411_label_6.png', 'id_1412_label_6.png', 'id_1413_label_7.png', 'id_1414_label_7.png', 'id_1415_label_8.png', 'id_1416_label_8.png', 'id_1417_label_9.png', 'id_1418_label_9.png', 'id_1419_label_10.png', 'id_1420_label_10.png', 'id_1421_label_11.png', 'id_1422_label_11.png', 'id_1423_label_12.png', 'id_1424_label_12.png', 'id_1425_label_13.png', 'id_1426_label_13.png', 'id_1427_label_14.png', 'id_1428_label_14.png', 'id_1429_label_15.png', 'id_1430_label_15.png', 'id_1431_label_16.png', 'id_1432_label_16.png', 'id_1433_label_17.png', 'id_1434_label_17.png', 'id_1435_label_18.png', 'id_1436_label_18.png', 'id_1437_label_19.png', 'id_1438_label_19.png', 'id_1439_label_20.png', 'id_1440_label_20.png', 'id_1441_label_21.png', 'id_1442_label_21.png', 'id_1443_label_22.png', 'id_1444_label_22.png', 'id_1445_label_23.png', 'id_1446_label_23.png', 'id_1447_label_24.png', 'id_1448_label_24.png', 'id_1449_label_25.png', 'id_1450_label_25.png', 'id_1451_label_26.png', 'id_1452_label_26.png', 'id_1453_label_27.png', 'id_1454_label_27.png', 'id_1455_label_28.png', 'id_1456_label_28.png', 'id_1457_label_1.png', 'id_1458_label_1.png', 'id_1459_label_2.png', 'id_1460_label_2.png', 'id_1461_label_3.png', 'id_1462_label_3.png', 'id_1463_label_4.png', 'id_1464_label_4.png', 'id_1465_label_5.png', 'id_1466_label_5.png', 'id_1467_label_6.png', 'id_1468_label_6.png', 'id_1469_label_7.png', 'id_1470_label_7.png', 'id_1471_label_8.png', 'id_1472_label_8.png', 'id_1473_label_9.png', 'id_1474_label_9.png', 'id_1475_label_10.png', 'id_1476_label_10.png', 'id_1477_label_11.png', 'id_1478_label_11.png', 'id_1479_label_12.png', 'id_1480_label_12.png', 'id_1481_label_13.png', 'id_1482_label_13.png', 'id_1483_label_14.png', 'id_1484_label_14.png', 'id_1485_label_15.png', 'id_1486_label_15.png', 'id_1487_label_16.png', 'id_1488_label_16.png', 'id_1489_label_17.png', 'id_1490_label_17.png', 'id_1491_label_18.png', 'id_1492_label_18.png', 'id_1493_label_19.png', 'id_1494_label_19.png', 'id_1495_label_20.png', 'id_1496_label_20.png', 'id_1497_label_21.png', 'id_1498_label_21.png', 'id_1499_label_22.png', 'id_1500_label_22.png', 'id_1501_label_23.png', 'id_1502_label_23.png', 'id_1503_label_24.png', 'id_1504_label_24.png', 'id_1505_label_25.png', 'id_1506_label_25.png', 'id_1507_label_26.png', 'id_1508_label_26.png', 'id_1509_label_27.png', 'id_1510_label_27.png', 'id_1511_label_28.png', 'id_1512_label_28.png', 'id_1513_label_1.png', 'id_1514_label_1.png', 'id_1515_label_2.png', 'id_1516_label_2.png', 'id_1517_label_3.png', 'id_1518_label_3.png', 'id_1519_label_4.png', 'id_1520_label_4.png', 'id_1521_label_5.png', 'id_1522_label_5.png', 'id_1523_label_6.png', 'id_1524_label_6.png', 'id_1525_label_7.png', 'id_1526_label_7.png', 'id_1527_label_8.png', 'id_1528_label_8.png', 'id_1529_label_9.png', 'id_1530_label_9.png', 'id_1531_label_10.png', 'id_1532_label_10.png', 'id_1533_label_11.png', 'id_1534_label_11.png', 'id_1535_label_12.png', 'id_1536_label_12.png', 'id_1537_label_13.png', 'id_1538_label_13.png', 'id_1539_label_14.png', 'id_1540_label_14.png', 'id_1541_label_15.png', 'id_1542_label_15.png', 'id_1543_label_16.png', 'id_1544_label_16.png', 'id_1545_label_17.png', 'id_1546_label_17.png', 'id_1547_label_18.png', 'id_1548_label_18.png', 'id_1549_label_19.png', 'id_1550_label_19.png', 'id_1551_label_20.png', 'id_1552_label_20.png', 'id_1553_label_21.png', 'id_1554_label_21.png', 'id_1555_label_22.png', 'id_1556_label_22.png', 'id_1557_label_23.png', 'id_1558_label_23.png', 'id_1559_label_24.png', 'id_1560_label_24.png', 'id_1561_label_25.png', 'id_1562_label_25.png', 'id_1563_label_26.png', 'id_1564_label_26.png', 'id_1565_label_27.png', 'id_1566_label_27.png', 'id_1567_label_28.png', 'id_1568_label_28.png', 'id_1569_label_1.png', 'id_1570_label_1.png', 'id_1571_label_2.png', 'id_1572_label_2.png', 'id_1573_label_3.png', 'id_1574_label_3.png', 'id_1575_label_4.png', 'id_1576_label_4.png', 'id_1577_label_5.png', 'id_1578_label_5.png', 'id_1579_label_6.png', 'id_1580_label_6.png', 'id_1581_label_7.png', 'id_1582_label_7.png', 'id_1583_label_8.png', 'id_1584_label_8.png', 'id_1585_label_9.png', 'id_1586_label_9.png', 'id_1587_label_10.png', 'id_1588_label_10.png', 'id_1589_label_11.png', 'id_1590_label_11.png', 'id_1591_label_12.png', 'id_1592_label_12.png', 'id_1593_label_13.png', 'id_1594_label_13.png', 'id_1595_label_14.png', 'id_1596_label_14.png', 'id_1597_label_15.png', 'id_1598_label_15.png', 'id_1599_label_16.png', 'id_1600_label_16.png', 'id_1601_label_17.png', 'id_1602_label_17.png', 'id_1603_label_18.png', 'id_1604_label_18.png', 'id_1605_label_19.png', 'id_1606_label_19.png', 'id_1607_label_20.png', 'id_1608_label_20.png', 'id_1609_label_21.png', 'id_1610_label_21.png', 'id_1611_label_22.png', 'id_1612_label_22.png', 'id_1613_label_23.png', 'id_1614_label_23.png', 'id_1615_label_24.png', 'id_1616_label_24.png', 'id_1617_label_25.png', 'id_1618_label_25.png', 'id_1619_label_26.png', 'id_1620_label_26.png', 'id_1621_label_27.png', 'id_1622_label_27.png', 'id_1623_label_28.png', 'id_1624_label_28.png', 'id_1625_label_1.png', 'id_1626_label_1.png', 'id_1627_label_2.png', 'id_1628_label_2.png', 'id_1629_label_3.png', 'id_1630_label_3.png', 'id_1631_label_4.png', 'id_1632_label_4.png', 'id_1633_label_5.png', 'id_1634_label_5.png', 'id_1635_label_6.png', 'id_1636_label_6.png', 'id_1637_label_7.png', 'id_1638_label_7.png', 'id_1639_label_8.png', 'id_1640_label_8.png', 'id_1641_label_9.png', 'id_1642_label_9.png', 'id_1643_label_10.png', 'id_1644_label_10.png', 'id_1645_label_11.png', 'id_1646_label_11.png', 'id_1647_label_12.png', 'id_1648_label_12.png', 'id_1649_label_13.png', 'id_1650_label_13.png', 'id_1651_label_14.png', 'id_1652_label_14.png', 'id_1653_label_15.png', 'id_1654_label_15.png', 'id_1655_label_16.png', 'id_1656_label_16.png', 'id_1657_label_17.png', 'id_1658_label_17.png', 'id_1659_label_18.png', 'id_1660_label_18.png', 'id_1661_label_19.png', 'id_1662_label_19.png', 'id_1663_label_20.png', 'id_1664_label_20.png', 'id_1665_label_21.png', 'id_1666_label_21.png', 'id_1667_label_22.png', 'id_1668_label_22.png', 'id_1669_label_23.png', 'id_1670_label_23.png', 'id_1671_label_24.png', 'id_1672_label_24.png', 'id_1673_label_25.png', 'id_1674_label_25.png', 'id_1675_label_26.png', 'id_1676_label_26.png', 'id_1677_label_27.png', 'id_1678_label_27.png', 'id_1679_label_28.png', 'id_1680_label_28.png', 'id_1681_label_1.png', 'id_1682_label_1.png', 'id_1683_label_2.png', 'id_1684_label_2.png', 'id_1685_label_3.png', 'id_1686_label_3.png', 'id_1687_label_4.png', 'id_1688_label_4.png', 'id_1689_label_5.png', 'id_1690_label_5.png', 'id_1691_label_6.png', 'id_1692_label_6.png', 'id_1693_label_7.png', 'id_1694_label_7.png', 'id_1695_label_8.png', 'id_1696_label_8.png', 'id_1697_label_9.png', 'id_1698_label_9.png', 'id_1699_label_10.png', 'id_1700_label_10.png', 'id_1701_label_11.png', 'id_1702_label_11.png', 'id_1703_label_12.png', 'id_1704_label_12.png', 'id_1705_label_13.png', 'id_1706_label_13.png', 'id_1707_label_14.png', 'id_1708_label_14.png', 'id_1709_label_15.png', 'id_1710_label_15.png', 'id_1711_label_16.png', 'id_1712_label_16.png', 'id_1713_label_17.png', 'id_1714_label_17.png', 'id_1715_label_18.png', 'id_1716_label_18.png', 'id_1717_label_19.png', 'id_1718_label_19.png', 'id_1719_label_20.png', 'id_1720_label_20.png', 'id_1721_label_21.png', 'id_1722_label_21.png', 'id_1723_label_22.png', 'id_1724_label_22.png', 'id_1725_label_23.png', 'id_1726_label_23.png', 'id_1727_label_24.png', 'id_1728_label_24.png', 'id_1729_label_25.png', 'id_1730_label_25.png', 'id_1731_label_26.png', 'id_1732_label_26.png', 'id_1733_label_27.png', 'id_1734_label_27.png', 'id_1735_label_28.png', 'id_1736_label_28.png', 'id_1737_label_1.png', 'id_1738_label_1.png', 'id_1739_label_2.png', 'id_1740_label_2.png', 'id_1741_label_3.png', 'id_1742_label_3.png', 'id_1743_label_4.png', 'id_1744_label_4.png', 'id_1745_label_5.png', 'id_1746_label_5.png', 'id_1747_label_6.png', 'id_1748_label_6.png', 'id_1749_label_7.png', 'id_1750_label_7.png', 'id_1751_label_8.png', 'id_1752_label_8.png', 'id_1753_label_9.png', 'id_1754_label_9.png', 'id_1755_label_10.png', 'id_1756_label_10.png', 'id_1757_label_11.png', 'id_1758_label_11.png', 'id_1759_label_12.png', 'id_1760_label_12.png', 'id_1761_label_13.png', 'id_1762_label_13.png', 'id_1763_label_14.png', 'id_1764_label_14.png', 'id_1765_label_15.png', 'id_1766_label_15.png', 'id_1767_label_16.png', 'id_1768_label_16.png', 'id_1769_label_17.png', 'id_1770_label_17.png', 'id_1771_label_18.png', 'id_1772_label_18.png', 'id_1773_label_19.png', 'id_1774_label_19.png', 'id_1775_label_20.png', 'id_1776_label_20.png', 'id_1777_label_21.png', 'id_1778_label_21.png', 'id_1779_label_22.png', 'id_1780_label_22.png', 'id_1781_label_23.png', 'id_1782_label_23.png', 'id_1783_label_24.png', 'id_1784_label_24.png', 'id_1785_label_25.png', 'id_1786_label_25.png', 'id_1787_label_26.png', 'id_1788_label_26.png', 'id_1789_label_27.png', 'id_1790_label_27.png', 'id_1791_label_28.png', 'id_1792_label_28.png', 'id_1793_label_1.png', 'id_1794_label_1.png', 'id_1795_label_2.png', 'id_1796_label_2.png', 'id_1797_label_3.png', 'id_1798_label_3.png', 'id_1799_label_4.png', 'id_1800_label_4.png', 'id_1801_label_5.png', 'id_1802_label_5.png', 'id_1803_label_6.png', 'id_1804_label_6.png', 'id_1805_label_7.png', 'id_1806_label_7.png', 'id_1807_label_8.png', 'id_1808_label_8.png', 'id_1809_label_9.png', 'id_1810_label_9.png', 'id_1811_label_10.png', 'id_1812_label_10.png', 'id_1813_label_11.png', 'id_1814_label_11.png', 'id_1815_label_12.png', 'id_1816_label_12.png', 'id_1817_label_13.png', 'id_1818_label_13.png', 'id_1819_label_14.png', 'id_1820_label_14.png', 'id_1821_label_15.png', 'id_1822_label_15.png', 'id_1823_label_16.png', 'id_1824_label_16.png', 'id_1825_label_17.png', 'id_1826_label_17.png', 'id_1827_label_18.png', 'id_1828_label_18.png', 'id_1829_label_19.png', 'id_1830_label_19.png', 'id_1831_label_20.png', 'id_1832_label_20.png', 'id_1833_label_21.png', 'id_1834_label_21.png', 'id_1835_label_22.png', 'id_1836_label_22.png', 'id_1837_label_23.png', 'id_1838_label_23.png', 'id_1839_label_24.png', 'id_1840_label_24.png', 'id_1841_label_25.png', 'id_1842_label_25.png', 'id_1843_label_26.png', 'id_1844_label_26.png', 'id_1845_label_27.png', 'id_1846_label_27.png', 'id_1847_label_28.png', 'id_1848_label_28.png', 'id_1849_label_1.png', 'id_1850_label_1.png', 'id_1851_label_2.png', 'id_1852_label_2.png', 'id_1853_label_3.png', 'id_1854_label_3.png', 'id_1855_label_4.png', 'id_1856_label_4.png', 'id_1857_label_5.png', 'id_1858_label_5.png', 'id_1859_label_6.png', 'id_1860_label_6.png', 'id_1861_label_7.png', 'id_1862_label_7.png', 'id_1863_label_8.png', 'id_1864_label_8.png', 'id_1865_label_9.png', 'id_1866_label_9.png', 'id_1867_label_10.png', 'id_1868_label_10.png', 'id_1869_label_11.png', 'id_1870_label_11.png', 'id_1871_label_12.png', 'id_1872_label_12.png', 'id_1873_label_13.png', 'id_1874_label_13.png', 'id_1875_label_14.png', 'id_1876_label_14.png', 'id_1877_label_15.png', 'id_1878_label_15.png', 'id_1879_label_16.png', 'id_1880_label_16.png', 'id_1881_label_17.png', 'id_1882_label_17.png', 'id_1883_label_18.png', 'id_1884_label_18.png', 'id_1885_label_19.png', 'id_1886_label_19.png', 'id_1887_label_20.png', 'id_1888_label_20.png', 'id_1889_label_21.png', 'id_1890_label_21.png', 'id_1891_label_22.png', 'id_1892_label_22.png', 'id_1893_label_23.png', 'id_1894_label_23.png', 'id_1895_label_24.png', 'id_1896_label_24.png', 'id_1897_label_25.png', 'id_1898_label_25.png', 'id_1899_label_26.png', 'id_1900_label_26.png', 'id_1901_label_27.png', 'id_1902_label_27.png', 'id_1903_label_28.png', 'id_1904_label_28.png', 'id_1905_label_1.png', 'id_1906_label_1.png', 'id_1907_label_2.png', 'id_1908_label_2.png', 'id_1909_label_3.png', 'id_1910_label_3.png', 'id_1911_label_4.png', 'id_1912_label_4.png', 'id_1913_label_5.png', 'id_1914_label_5.png', 'id_1915_label_6.png', 'id_1916_label_6.png', 'id_1917_label_7.png', 'id_1918_label_7.png', 'id_1919_label_8.png', 'id_1920_label_8.png', 'id_1921_label_9.png', 'id_1922_label_9.png', 'id_1923_label_10.png', 'id_1924_label_10.png', 'id_1925_label_11.png', 'id_1926_label_11.png', 'id_1927_label_12.png', 'id_1928_label_12.png', 'id_1929_label_13.png', 'id_1930_label_13.png', 'id_1931_label_14.png', 'id_1932_label_14.png', 'id_1933_label_15.png', 'id_1934_label_15.png', 'id_1935_label_16.png', 'id_1936_label_16.png', 'id_1937_label_17.png', 'id_1938_label_17.png', 'id_1939_label_18.png', 'id_1940_label_18.png', 'id_1941_label_19.png', 'id_1942_label_19.png', 'id_1943_label_20.png', 'id_1944_label_20.png', 'id_1945_label_21.png', 'id_1946_label_21.png', 'id_1947_label_22.png', 'id_1948_label_22.png', 'id_1949_label_23.png', 'id_1950_label_23.png', 'id_1951_label_24.png', 'id_1952_label_24.png', 'id_1953_label_25.png', 'id_1954_label_25.png', 'id_1955_label_26.png', 'id_1956_label_26.png', 'id_1957_label_27.png', 'id_1958_label_27.png', 'id_1959_label_28.png', 'id_1960_label_28.png', 'id_1961_label_1.png', 'id_1962_label_1.png', 'id_1963_label_2.png', 'id_1964_label_2.png', 'id_1965_label_3.png', 'id_1966_label_3.png', 'id_1967_label_4.png', 'id_1968_label_4.png', 'id_1969_label_5.png', 'id_1970_label_5.png', 'id_1971_label_6.png', 'id_1972_label_6.png', 'id_1973_label_7.png', 'id_1974_label_7.png', 'id_1975_label_8.png', 'id_1976_label_8.png', 'id_1977_label_9.png', 'id_1978_label_9.png', 'id_1979_label_10.png', 'id_1980_label_10.png', 'id_1981_label_11.png', 'id_1982_label_11.png', 'id_1983_label_12.png', 'id_1984_label_12.png', 'id_1985_label_13.png', 'id_1986_label_13.png', 'id_1987_label_14.png', 'id_1988_label_14.png', 'id_1989_label_15.png', 'id_1990_label_15.png', 'id_1991_label_16.png', 'id_1992_label_16.png', 'id_1993_label_17.png', 'id_1994_label_17.png', 'id_1995_label_18.png', 'id_1996_label_18.png', 'id_1997_label_19.png', 'id_1998_label_19.png', 'id_1999_label_20.png', 'id_2000_label_20.png', 'id_2001_label_21.png', 'id_2002_label_21.png', 'id_2003_label_22.png', 'id_2004_label_22.png', 'id_2005_label_23.png', 'id_2006_label_23.png', 'id_2007_label_24.png', 'id_2008_label_24.png', 'id_2009_label_25.png', 'id_2010_label_25.png', 'id_2011_label_26.png', 'id_2012_label_26.png', 'id_2013_label_27.png', 'id_2014_label_27.png', 'id_2015_label_28.png', 'id_2016_label_28.png', 'id_2017_label_1.png', 'id_2018_label_1.png', 'id_2019_label_2.png', 'id_2020_label_2.png', 'id_2021_label_3.png', 'id_2022_label_3.png', 'id_2023_label_4.png', 'id_2024_label_4.png', 'id_2025_label_5.png', 'id_2026_label_5.png', 'id_2027_label_6.png', 'id_2028_label_6.png', 'id_2029_label_7.png', 'id_2030_label_7.png', 'id_2031_label_8.png', 'id_2032_label_8.png', 'id_2033_label_9.png', 'id_2034_label_9.png', 'id_2035_label_10.png', 'id_2036_label_10.png', 'id_2037_label_11.png', 'id_2038_label_11.png', 'id_2039_label_12.png', 'id_2040_label_12.png', 'id_2041_label_13.png', 'id_2042_label_13.png', 'id_2043_label_14.png', 'id_2044_label_14.png', 'id_2045_label_15.png', 'id_2046_label_15.png', 'id_2047_label_16.png', 'id_2048_label_16.png', 'id_2049_label_17.png', 'id_2050_label_17.png', 'id_2051_label_18.png', 'id_2052_label_18.png', 'id_2053_label_19.png', 'id_2054_label_19.png', 'id_2055_label_20.png', 'id_2056_label_20.png', 'id_2057_label_21.png', 'id_2058_label_21.png', 'id_2059_label_22.png', 'id_2060_label_22.png', 'id_2061_label_23.png', 'id_2062_label_23.png', 'id_2063_label_24.png', 'id_2064_label_24.png', 'id_2065_label_25.png', 'id_2066_label_25.png', 'id_2067_label_26.png', 'id_2068_label_26.png', 'id_2069_label_27.png', 'id_2070_label_27.png', 'id_2071_label_28.png', 'id_2072_label_28.png', 'id_2073_label_1.png', 'id_2074_label_1.png', 'id_2075_label_2.png', 'id_2076_label_2.png', 'id_2077_label_3.png', 'id_2078_label_3.png', 'id_2079_label_4.png', 'id_2080_label_4.png', 'id_2081_label_5.png', 'id_2082_label_5.png', 'id_2083_label_6.png', 'id_2084_label_6.png', 'id_2085_label_7.png', 'id_2086_label_7.png', 'id_2087_label_8.png', 'id_2088_label_8.png', 'id_2089_label_9.png', 'id_2090_label_9.png', 'id_2091_label_10.png', 'id_2092_label_10.png', 'id_2093_label_11.png', 'id_2094_label_11.png', 'id_2095_label_12.png', 'id_2096_label_12.png', 'id_2097_label_13.png', 'id_2098_label_13.png', 'id_2099_label_14.png', 'id_2100_label_14.png', 'id_2101_label_15.png', 'id_2102_label_15.png', 'id_2103_label_16.png', 'id_2104_label_16.png', 'id_2105_label_17.png', 'id_2106_label_17.png', 'id_2107_label_18.png', 'id_2108_label_18.png', 'id_2109_label_19.png', 'id_2110_label_19.png', 'id_2111_label_20.png', 'id_2112_label_20.png', 'id_2113_label_21.png', 'id_2114_label_21.png', 'id_2115_label_22.png', 'id_2116_label_22.png', 'id_2117_label_23.png', 'id_2118_label_23.png', 'id_2119_label_24.png', 'id_2120_label_24.png', 'id_2121_label_25.png', 'id_2122_label_25.png', 'id_2123_label_26.png', 'id_2124_label_26.png', 'id_2125_label_27.png', 'id_2126_label_27.png', 'id_2127_label_28.png', 'id_2128_label_28.png', 'id_2129_label_1.png', 'id_2130_label_1.png', 'id_2131_label_2.png', 'id_2132_label_2.png', 'id_2133_label_3.png', 'id_2134_label_3.png', 'id_2135_label_4.png', 'id_2136_label_4.png', 'id_2137_label_5.png', 'id_2138_label_5.png', 'id_2139_label_6.png', 'id_2140_label_6.png', 'id_2141_label_7.png', 'id_2142_label_7.png', 'id_2143_label_8.png', 'id_2144_label_8.png', 'id_2145_label_9.png', 'id_2146_label_9.png', 'id_2147_label_10.png', 'id_2148_label_10.png', 'id_2149_label_11.png', 'id_2150_label_11.png', 'id_2151_label_12.png', 'id_2152_label_12.png', 'id_2153_label_13.png', 'id_2154_label_13.png', 'id_2155_label_14.png', 'id_2156_label_14.png', 'id_2157_label_15.png', 'id_2158_label_15.png', 'id_2159_label_16.png', 'id_2160_label_16.png', 'id_2161_label_17.png', 'id_2162_label_17.png', 'id_2163_label_18.png', 'id_2164_label_18.png', 'id_2165_label_19.png', 'id_2166_label_19.png', 'id_2167_label_20.png', 'id_2168_label_20.png', 'id_2169_label_21.png', 'id_2170_label_21.png', 'id_2171_label_22.png', 'id_2172_label_22.png', 'id_2173_label_23.png', 'id_2174_label_23.png', 'id_2175_label_24.png', 'id_2176_label_24.png', 'id_2177_label_25.png', 'id_2178_label_25.png', 'id_2179_label_26.png', 'id_2180_label_26.png', 'id_2181_label_27.png', 'id_2182_label_27.png', 'id_2183_label_28.png', 'id_2184_label_28.png', 'id_2185_label_1.png', 'id_2186_label_1.png', 'id_2187_label_2.png', 'id_2188_label_2.png', 'id_2189_label_3.png', 'id_2190_label_3.png', 'id_2191_label_4.png', 'id_2192_label_4.png', 'id_2193_label_5.png', 'id_2194_label_5.png', 'id_2195_label_6.png', 'id_2196_label_6.png', 'id_2197_label_7.png', 'id_2198_label_7.png', 'id_2199_label_8.png', 'id_2200_label_8.png', 'id_2201_label_9.png', 'id_2202_label_9.png', 'id_2203_label_10.png', 'id_2204_label_10.png', 'id_2205_label_11.png', 'id_2206_label_11.png', 'id_2207_label_12.png', 'id_2208_label_12.png', 'id_2209_label_13.png', 'id_2210_label_13.png', 'id_2211_label_14.png', 'id_2212_label_14.png', 'id_2213_label_15.png', 'id_2214_label_15.png', 'id_2215_label_16.png', 'id_2216_label_16.png', 'id_2217_label_17.png', 'id_2218_label_17.png', 'id_2219_label_18.png', 'id_2220_label_18.png', 'id_2221_label_19.png', 'id_2222_label_19.png', 'id_2223_label_20.png', 'id_2224_label_20.png', 'id_2225_label_21.png', 'id_2226_label_21.png', 'id_2227_label_22.png', 'id_2228_label_22.png', 'id_2229_label_23.png', 'id_2230_label_23.png', 'id_2231_label_24.png', 'id_2232_label_24.png', 'id_2233_label_25.png', 'id_2234_label_25.png', 'id_2235_label_26.png', 'id_2236_label_26.png', 'id_2237_label_27.png', 'id_2238_label_27.png', 'id_2239_label_28.png', 'id_2240_label_28.png', 'id_2241_label_1.png', 'id_2242_label_1.png', 'id_2243_label_2.png', 'id_2244_label_2.png', 'id_2245_label_3.png', 'id_2246_label_3.png', 'id_2247_label_4.png', 'id_2248_label_4.png', 'id_2249_label_5.png', 'id_2250_label_5.png', 'id_2251_label_6.png', 'id_2252_label_6.png', 'id_2253_label_7.png', 'id_2254_label_7.png', 'id_2255_label_8.png', 'id_2256_label_8.png', 'id_2257_label_9.png', 'id_2258_label_9.png', 'id_2259_label_10.png', 'id_2260_label_10.png', 'id_2261_label_11.png', 'id_2262_label_11.png', 'id_2263_label_12.png', 'id_2264_label_12.png', 'id_2265_label_13.png', 'id_2266_label_13.png', 'id_2267_label_14.png', 'id_2268_label_14.png', 'id_2269_label_15.png', 'id_2270_label_15.png', 'id_2271_label_16.png', 'id_2272_label_16.png', 'id_2273_label_17.png', 'id_2274_label_17.png', 'id_2275_label_18.png', 'id_2276_label_18.png', 'id_2277_label_19.png', 'id_2278_label_19.png', 'id_2279_label_20.png', 'id_2280_label_20.png', 'id_2281_label_21.png', 'id_2282_label_21.png', 'id_2283_label_22.png', 'id_2284_label_22.png', 'id_2285_label_23.png', 'id_2286_label_23.png', 'id_2287_label_24.png', 'id_2288_label_24.png', 'id_2289_label_25.png', 'id_2290_label_25.png', 'id_2291_label_26.png', 'id_2292_label_26.png', 'id_2293_label_27.png', 'id_2294_label_27.png', 'id_2295_label_28.png', 'id_2296_label_28.png', 'id_2297_label_1.png', 'id_2298_label_1.png', 'id_2299_label_2.png', 'id_2300_label_2.png', 'id_2301_label_3.png', 'id_2302_label_3.png', 'id_2303_label_4.png', 'id_2304_label_4.png', 'id_2305_label_5.png', 'id_2306_label_5.png', 'id_2307_label_6.png', 'id_2308_label_6.png', 'id_2309_label_7.png', 'id_2310_label_7.png', 'id_2311_label_8.png', 'id_2312_label_8.png', 'id_2313_label_9.png', 'id_2314_label_9.png', 'id_2315_label_10.png', 'id_2316_label_10.png', 'id_2317_label_11.png', 'id_2318_label_11.png', 'id_2319_label_12.png', 'id_2320_label_12.png', 'id_2321_label_13.png', 'id_2322_label_13.png', 'id_2323_label_14.png', 'id_2324_label_14.png', 'id_2325_label_15.png', 'id_2326_label_15.png', 'id_2327_label_16.png', 'id_2328_label_16.png', 'id_2329_label_17.png', 'id_2330_label_17.png', 'id_2331_label_18.png', 'id_2332_label_18.png', 'id_2333_label_19.png', 'id_2334_label_19.png', 'id_2335_label_20.png', 'id_2336_label_20.png', 'id_2337_label_21.png', 'id_2338_label_21.png', 'id_2339_label_22.png', 'id_2340_label_22.png', 'id_2341_label_23.png', 'id_2342_label_23.png', 'id_2343_label_24.png', 'id_2344_label_24.png', 'id_2345_label_25.png', 'id_2346_label_25.png', 'id_2347_label_26.png', 'id_2348_label_26.png', 'id_2349_label_27.png', 'id_2350_label_27.png', 'id_2351_label_28.png', 'id_2352_label_28.png', 'id_2353_label_1.png', 'id_2354_label_1.png', 'id_2355_label_2.png', 'id_2356_label_2.png', 'id_2357_label_3.png', 'id_2358_label_3.png', 'id_2359_label_4.png', 'id_2360_label_4.png', 'id_2361_label_5.png', 'id_2362_label_5.png', 'id_2363_label_6.png', 'id_2364_label_6.png', 'id_2365_label_7.png', 'id_2366_label_7.png', 'id_2367_label_8.png', 'id_2368_label_8.png', 'id_2369_label_9.png', 'id_2370_label_9.png', 'id_2371_label_10.png', 'id_2372_label_10.png', 'id_2373_label_11.png', 'id_2374_label_11.png', 'id_2375_label_12.png', 'id_2376_label_12.png', 'id_2377_label_13.png', 'id_2378_label_13.png', 'id_2379_label_14.png', 'id_2380_label_14.png', 'id_2381_label_15.png', 'id_2382_label_15.png', 'id_2383_label_16.png', 'id_2384_label_16.png', 'id_2385_label_17.png', 'id_2386_label_17.png', 'id_2387_label_18.png', 'id_2388_label_18.png', 'id_2389_label_19.png', 'id_2390_label_19.png', 'id_2391_label_20.png', 'id_2392_label_20.png', 'id_2393_label_21.png', 'id_2394_label_21.png', 'id_2395_label_22.png', 'id_2396_label_22.png', 'id_2397_label_23.png', 'id_2398_label_23.png', 'id_2399_label_24.png', 'id_2400_label_24.png', 'id_2401_label_25.png', 'id_2402_label_25.png', 'id_2403_label_26.png', 'id_2404_label_26.png', 'id_2405_label_27.png', 'id_2406_label_27.png', 'id_2407_label_28.png', 'id_2408_label_28.png', 'id_2409_label_1.png', 'id_2410_label_1.png', 'id_2411_label_2.png', 'id_2412_label_2.png', 'id_2413_label_3.png', 'id_2414_label_3.png', 'id_2415_label_4.png', 'id_2416_label_4.png', 'id_2417_label_5.png', 'id_2418_label_5.png', 'id_2419_label_6.png', 'id_2420_label_6.png', 'id_2421_label_7.png', 'id_2422_label_7.png', 'id_2423_label_8.png', 'id_2424_label_8.png', 'id_2425_label_9.png', 'id_2426_label_9.png', 'id_2427_label_10.png', 'id_2428_label_10.png', 'id_2429_label_11.png', 'id_2430_label_11.png', 'id_2431_label_12.png', 'id_2432_label_12.png', 'id_2433_label_13.png', 'id_2434_label_13.png', 'id_2435_label_14.png', 'id_2436_label_14.png', 'id_2437_label_15.png', 'id_2438_label_15.png', 'id_2439_label_16.png', 'id_2440_label_16.png', 'id_2441_label_17.png', 'id_2442_label_17.png', 'id_2443_label_18.png', 'id_2444_label_18.png', 'id_2445_label_19.png', 'id_2446_label_19.png', 'id_2447_label_20.png', 'id_2448_label_20.png', 'id_2449_label_21.png', 'id_2450_label_21.png', 'id_2451_label_22.png', 'id_2452_label_22.png', 'id_2453_label_23.png', 'id_2454_label_23.png', 'id_2455_label_24.png', 'id_2456_label_24.png', 'id_2457_label_25.png', 'id_2458_label_25.png', 'id_2459_label_26.png', 'id_2460_label_26.png', 'id_2461_label_27.png', 'id_2462_label_27.png', 'id_2463_label_28.png', 'id_2464_label_28.png', 'id_2465_label_1.png', 'id_2466_label_1.png', 'id_2467_label_2.png', 'id_2468_label_2.png', 'id_2469_label_3.png', 'id_2470_label_3.png', 'id_2471_label_4.png', 'id_2472_label_4.png', 'id_2473_label_5.png', 'id_2474_label_5.png', 'id_2475_label_6.png', 'id_2476_label_6.png', 'id_2477_label_7.png', 'id_2478_label_7.png', 'id_2479_label_8.png', 'id_2480_label_8.png', 'id_2481_label_9.png', 'id_2482_label_9.png', 'id_2483_label_10.png', 'id_2484_label_10.png', 'id_2485_label_11.png', 'id_2486_label_11.png', 'id_2487_label_12.png', 'id_2488_label_12.png', 'id_2489_label_13.png', 'id_2490_label_13.png', 'id_2491_label_14.png', 'id_2492_label_14.png', 'id_2493_label_15.png', 'id_2494_label_15.png', 'id_2495_label_16.png', 'id_2496_label_16.png', 'id_2497_label_17.png', 'id_2498_label_17.png', 'id_2499_label_18.png', 'id_2500_label_18.png', 'id_2501_label_19.png', 'id_2502_label_19.png', 'id_2503_label_20.png', 'id_2504_label_20.png', 'id_2505_label_21.png', 'id_2506_label_21.png', 'id_2507_label_22.png', 'id_2508_label_22.png', 'id_2509_label_23.png', 'id_2510_label_23.png', 'id_2511_label_24.png', 'id_2512_label_24.png', 'id_2513_label_25.png', 'id_2514_label_25.png', 'id_2515_label_26.png', 'id_2516_label_26.png', 'id_2517_label_27.png', 'id_2518_label_27.png', 'id_2519_label_28.png', 'id_2520_label_28.png', 'id_2521_label_1.png', 'id_2522_label_1.png', 'id_2523_label_2.png', 'id_2524_label_2.png', 'id_2525_label_3.png', 'id_2526_label_3.png', 'id_2527_label_4.png', 'id_2528_label_4.png', 'id_2529_label_5.png', 'id_2530_label_5.png', 'id_2531_label_6.png', 'id_2532_label_6.png', 'id_2533_label_7.png', 'id_2534_label_7.png', 'id_2535_label_8.png', 'id_2536_label_8.png', 'id_2537_label_9.png', 'id_2538_label_9.png', 'id_2539_label_10.png', 'id_2540_label_10.png', 'id_2541_label_11.png', 'id_2542_label_11.png', 'id_2543_label_12.png', 'id_2544_label_12.png', 'id_2545_label_13.png', 'id_2546_label_13.png', 'id_2547_label_14.png', 'id_2548_label_14.png', 'id_2549_label_15.png', 'id_2550_label_15.png', 'id_2551_label_16.png', 'id_2552_label_16.png', 'id_2553_label_17.png', 'id_2554_label_17.png', 'id_2555_label_18.png', 'id_2556_label_18.png', 'id_2557_label_19.png', 'id_2558_label_19.png', 'id_2559_label_20.png', 'id_2560_label_20.png', 'id_2561_label_21.png', 'id_2562_label_21.png', 'id_2563_label_22.png', 'id_2564_label_22.png', 'id_2565_label_23.png', 'id_2566_label_23.png', 'id_2567_label_24.png', 'id_2568_label_24.png', 'id_2569_label_25.png', 'id_2570_label_25.png', 'id_2571_label_26.png', 'id_2572_label_26.png', 'id_2573_label_27.png', 'id_2574_label_27.png', 'id_2575_label_28.png', 'id_2576_label_28.png', 'id_2577_label_1.png', 'id_2578_label_1.png', 'id_2579_label_2.png', 'id_2580_label_2.png', 'id_2581_label_3.png', 'id_2582_label_3.png', 'id_2583_label_4.png', 'id_2584_label_4.png', 'id_2585_label_5.png', 'id_2586_label_5.png', 'id_2587_label_6.png', 'id_2588_label_6.png', 'id_2589_label_7.png', 'id_2590_label_7.png', 'id_2591_label_8.png', 'id_2592_label_8.png', 'id_2593_label_9.png', 'id_2594_label_9.png', 'id_2595_label_10.png', 'id_2596_label_10.png', 'id_2597_label_11.png', 'id_2598_label_11.png', 'id_2599_label_12.png', 'id_2600_label_12.png', 'id_2601_label_13.png', 'id_2602_label_13.png', 'id_2603_label_14.png', 'id_2604_label_14.png', 'id_2605_label_15.png', 'id_2606_label_15.png', 'id_2607_label_16.png', 'id_2608_label_16.png', 'id_2609_label_17.png', 'id_2610_label_17.png', 'id_2611_label_18.png', 'id_2612_label_18.png', 'id_2613_label_19.png', 'id_2614_label_19.png', 'id_2615_label_20.png', 'id_2616_label_20.png', 'id_2617_label_21.png', 'id_2618_label_21.png', 'id_2619_label_22.png', 'id_2620_label_22.png', 'id_2621_label_23.png', 'id_2622_label_23.png', 'id_2623_label_24.png', 'id_2624_label_24.png', 'id_2625_label_25.png', 'id_2626_label_25.png', 'id_2627_label_26.png', 'id_2628_label_26.png', 'id_2629_label_27.png', 'id_2630_label_27.png', 'id_2631_label_28.png', 'id_2632_label_28.png', 'id_2633_label_1.png', 'id_2634_label_1.png', 'id_2635_label_2.png', 'id_2636_label_2.png', 'id_2637_label_3.png', 'id_2638_label_3.png', 'id_2639_label_4.png', 'id_2640_label_4.png', 'id_2641_label_5.png', 'id_2642_label_5.png', 'id_2643_label_6.png', 'id_2644_label_6.png', 'id_2645_label_7.png', 'id_2646_label_7.png', 'id_2647_label_8.png', 'id_2648_label_8.png', 'id_2649_label_9.png', 'id_2650_label_9.png', 'id_2651_label_10.png', 'id_2652_label_10.png', 'id_2653_label_11.png', 'id_2654_label_11.png', 'id_2655_label_12.png', 'id_2656_label_12.png', 'id_2657_label_13.png', 'id_2658_label_13.png', 'id_2659_label_14.png', 'id_2660_label_14.png', 'id_2661_label_15.png', 'id_2662_label_15.png', 'id_2663_label_16.png', 'id_2664_label_16.png', 'id_2665_label_17.png', 'id_2666_label_17.png', 'id_2667_label_18.png', 'id_2668_label_18.png', 'id_2669_label_19.png', 'id_2670_label_19.png', 'id_2671_label_20.png', 'id_2672_label_20.png', 'id_2673_label_21.png', 'id_2674_label_21.png', 'id_2675_label_22.png', 'id_2676_label_22.png', 'id_2677_label_23.png', 'id_2678_label_23.png', 'id_2679_label_24.png', 'id_2680_label_24.png', 'id_2681_label_25.png', 'id_2682_label_25.png', 'id_2683_label_26.png', 'id_2684_label_26.png', 'id_2685_label_27.png', 'id_2686_label_27.png', 'id_2687_label_28.png', 'id_2688_label_28.png', 'id_2689_label_1.png', 'id_2690_label_1.png', 'id_2691_label_2.png', 'id_2692_label_2.png', 'id_2693_label_3.png', 'id_2694_label_3.png', 'id_2695_label_4.png', 'id_2696_label_4.png', 'id_2697_label_5.png', 'id_2698_label_5.png', 'id_2699_label_6.png', 'id_2700_label_6.png', 'id_2701_label_7.png', 'id_2702_label_7.png', 'id_2703_label_8.png', 'id_2704_label_8.png', 'id_2705_label_9.png', 'id_2706_label_9.png', 'id_2707_label_10.png', 'id_2708_label_10.png', 'id_2709_label_11.png', 'id_2710_label_11.png', 'id_2711_label_12.png', 'id_2712_label_12.png', 'id_2713_label_13.png', 'id_2714_label_13.png', 'id_2715_label_14.png', 'id_2716_label_14.png', 'id_2717_label_15.png', 'id_2718_label_15.png', 'id_2719_label_16.png', 'id_2720_label_16.png', 'id_2721_label_17.png', 'id_2722_label_17.png', 'id_2723_label_18.png', 'id_2724_label_18.png', 'id_2725_label_19.png', 'id_2726_label_19.png', 'id_2727_label_20.png', 'id_2728_label_20.png', 'id_2729_label_21.png', 'id_2730_label_21.png', 'id_2731_label_22.png', 'id_2732_label_22.png', 'id_2733_label_23.png', 'id_2734_label_23.png', 'id_2735_label_24.png', 'id_2736_label_24.png', 'id_2737_label_25.png', 'id_2738_label_25.png', 'id_2739_label_26.png', 'id_2740_label_26.png', 'id_2741_label_27.png', 'id_2742_label_27.png', 'id_2743_label_28.png', 'id_2744_label_28.png', 'id_2745_label_1.png', 'id_2746_label_1.png', 'id_2747_label_2.png', 'id_2748_label_2.png', 'id_2749_label_3.png', 'id_2750_label_3.png', 'id_2751_label_4.png', 'id_2752_label_4.png', 'id_2753_label_5.png', 'id_2754_label_5.png', 'id_2755_label_6.png', 'id_2756_label_6.png', 'id_2757_label_7.png', 'id_2758_label_7.png', 'id_2759_label_8.png', 'id_2760_label_8.png', 'id_2761_label_9.png', 'id_2762_label_9.png', 'id_2763_label_10.png', 'id_2764_label_10.png', 'id_2765_label_11.png', 'id_2766_label_11.png', 'id_2767_label_12.png', 'id_2768_label_12.png', 'id_2769_label_13.png', 'id_2770_label_13.png', 'id_2771_label_14.png', 'id_2772_label_14.png', 'id_2773_label_15.png', 'id_2774_label_15.png', 'id_2775_label_16.png', 'id_2776_label_16.png', 'id_2777_label_17.png', 'id_2778_label_17.png', 'id_2779_label_18.png', 'id_2780_label_18.png', 'id_2781_label_19.png', 'id_2782_label_19.png', 'id_2783_label_20.png', 'id_2784_label_20.png', 'id_2785_label_21.png', 'id_2786_label_21.png', 'id_2787_label_22.png', 'id_2788_label_22.png', 'id_2789_label_23.png', 'id_2790_label_23.png', 'id_2791_label_24.png', 'id_2792_label_24.png', 'id_2793_label_25.png', 'id_2794_label_25.png', 'id_2795_label_26.png', 'id_2796_label_26.png', 'id_2797_label_27.png', 'id_2798_label_27.png', 'id_2799_label_28.png', 'id_2800_label_28.png', 'id_2801_label_1.png', 'id_2802_label_1.png', 'id_2803_label_2.png', 'id_2804_label_2.png', 'id_2805_label_3.png', 'id_2806_label_3.png', 'id_2807_label_4.png', 'id_2808_label_4.png', 'id_2809_label_5.png', 'id_2810_label_5.png', 'id_2811_label_6.png', 'id_2812_label_6.png', 'id_2813_label_7.png', 'id_2814_label_7.png', 'id_2815_label_8.png', 'id_2816_label_8.png', 'id_2817_label_9.png', 'id_2818_label_9.png', 'id_2819_label_10.png', 'id_2820_label_10.png', 'id_2821_label_11.png', 'id_2822_label_11.png', 'id_2823_label_12.png', 'id_2824_label_12.png', 'id_2825_label_13.png', 'id_2826_label_13.png', 'id_2827_label_14.png', 'id_2828_label_14.png', 'id_2829_label_15.png', 'id_2830_label_15.png', 'id_2831_label_16.png', 'id_2832_label_16.png', 'id_2833_label_17.png', 'id_2834_label_17.png', 'id_2835_label_18.png', 'id_2836_label_18.png', 'id_2837_label_19.png', 'id_2838_label_19.png', 'id_2839_label_20.png', 'id_2840_label_20.png', 'id_2841_label_21.png', 'id_2842_label_21.png', 'id_2843_label_22.png', 'id_2844_label_22.png', 'id_2845_label_23.png', 'id_2846_label_23.png', 'id_2847_label_24.png', 'id_2848_label_24.png', 'id_2849_label_25.png', 'id_2850_label_25.png', 'id_2851_label_26.png', 'id_2852_label_26.png', 'id_2853_label_27.png', 'id_2854_label_27.png', 'id_2855_label_28.png', 'id_2856_label_28.png', 'id_2857_label_1.png', 'id_2858_label_1.png', 'id_2859_label_2.png', 'id_2860_label_2.png', 'id_2861_label_3.png', 'id_2862_label_3.png', 'id_2863_label_4.png', 'id_2864_label_4.png', 'id_2865_label_5.png', 'id_2866_label_5.png', 'id_2867_label_6.png', 'id_2868_label_6.png', 'id_2869_label_7.png', 'id_2870_label_7.png', 'id_2871_label_8.png', 'id_2872_label_8.png', 'id_2873_label_9.png', 'id_2874_label_9.png', 'id_2875_label_10.png', 'id_2876_label_10.png', 'id_2877_label_11.png', 'id_2878_label_11.png', 'id_2879_label_12.png', 'id_2880_label_12.png', 'id_2881_label_13.png', 'id_2882_label_13.png', 'id_2883_label_14.png', 'id_2884_label_14.png', 'id_2885_label_15.png', 'id_2886_label_15.png', 'id_2887_label_16.png', 'id_2888_label_16.png', 'id_2889_label_17.png', 'id_2890_label_17.png', 'id_2891_label_18.png', 'id_2892_label_18.png', 'id_2893_label_19.png', 'id_2894_label_19.png', 'id_2895_label_20.png', 'id_2896_label_20.png', 'id_2897_label_21.png', 'id_2898_label_21.png', 'id_2899_label_22.png', 'id_2900_label_22.png', 'id_2901_label_23.png', 'id_2902_label_23.png', 'id_2903_label_24.png', 'id_2904_label_24.png', 'id_2905_label_25.png', 'id_2906_label_25.png', 'id_2907_label_26.png', 'id_2908_label_26.png', 'id_2909_label_27.png', 'id_2910_label_27.png', 'id_2911_label_28.png', 'id_2912_label_28.png', 'id_2913_label_1.png', 'id_2914_label_1.png', 'id_2915_label_2.png', 'id_2916_label_2.png', 'id_2917_label_3.png', 'id_2918_label_3.png', 'id_2919_label_4.png', 'id_2920_label_4.png', 'id_2921_label_5.png', 'id_2922_label_5.png', 'id_2923_label_6.png', 'id_2924_label_6.png', 'id_2925_label_7.png', 'id_2926_label_7.png', 'id_2927_label_8.png', 'id_2928_label_8.png', 'id_2929_label_9.png', 'id_2930_label_9.png', 'id_2931_label_10.png', 'id_2932_label_10.png', 'id_2933_label_11.png', 'id_2934_label_11.png', 'id_2935_label_12.png', 'id_2936_label_12.png', 'id_2937_label_13.png', 'id_2938_label_13.png', 'id_2939_label_14.png', 'id_2940_label_14.png', 'id_2941_label_15.png', 'id_2942_label_15.png', 'id_2943_label_16.png', 'id_2944_label_16.png', 'id_2945_label_17.png', 'id_2946_label_17.png', 'id_2947_label_18.png', 'id_2948_label_18.png', 'id_2949_label_19.png', 'id_2950_label_19.png', 'id_2951_label_20.png', 'id_2952_label_20.png', 'id_2953_label_21.png', 'id_2954_label_21.png', 'id_2955_label_22.png', 'id_2956_label_22.png', 'id_2957_label_23.png', 'id_2958_label_23.png', 'id_2959_label_24.png', 'id_2960_label_24.png', 'id_2961_label_25.png', 'id_2962_label_25.png', 'id_2963_label_26.png', 'id_2964_label_26.png', 'id_2965_label_27.png', 'id_2966_label_27.png', 'id_2967_label_28.png', 'id_2968_label_28.png', 'id_2969_label_1.png', 'id_2970_label_1.png', 'id_2971_label_2.png', 'id_2972_label_2.png', 'id_2973_label_3.png', 'id_2974_label_3.png', 'id_2975_label_4.png', 'id_2976_label_4.png', 'id_2977_label_5.png', 'id_2978_label_5.png', 'id_2979_label_6.png', 'id_2980_label_6.png', 'id_2981_label_7.png', 'id_2982_label_7.png', 'id_2983_label_8.png', 'id_2984_label_8.png', 'id_2985_label_9.png', 'id_2986_label_9.png', 'id_2987_label_10.png', 'id_2988_label_10.png', 'id_2989_label_11.png', 'id_2990_label_11.png', 'id_2991_label_12.png', 'id_2992_label_12.png', 'id_2993_label_13.png', 'id_2994_label_13.png', 'id_2995_label_14.png', 'id_2996_label_14.png', 'id_2997_label_15.png', 'id_2998_label_15.png', 'id_2999_label_16.png', 'id_3000_label_16.png', 'id_3001_label_17.png', 'id_3002_label_17.png', 'id_3003_label_18.png', 'id_3004_label_18.png', 'id_3005_label_19.png', 'id_3006_label_19.png', 'id_3007_label_20.png', 'id_3008_label_20.png', 'id_3009_label_21.png', 'id_3010_label_21.png', 'id_3011_label_22.png', 'id_3012_label_22.png', 'id_3013_label_23.png', 'id_3014_label_23.png', 'id_3015_label_24.png', 'id_3016_label_24.png', 'id_3017_label_25.png', 'id_3018_label_25.png', 'id_3019_label_26.png', 'id_3020_label_26.png', 'id_3021_label_27.png', 'id_3022_label_27.png', 'id_3023_label_28.png', 'id_3024_label_28.png', 'id_3025_label_1.png', 'id_3026_label_1.png', 'id_3027_label_2.png', 'id_3028_label_2.png', 'id_3029_label_3.png', 'id_3030_label_3.png', 'id_3031_label_4.png', 'id_3032_label_4.png', 'id_3033_label_5.png', 'id_3034_label_5.png', 'id_3035_label_6.png', 'id_3036_label_6.png', 'id_3037_label_7.png', 'id_3038_label_7.png', 'id_3039_label_8.png', 'id_3040_label_8.png', 'id_3041_label_9.png', 'id_3042_label_9.png', 'id_3043_label_10.png', 'id_3044_label_10.png', 'id_3045_label_11.png', 'id_3046_label_11.png', 'id_3047_label_12.png', 'id_3048_label_12.png', 'id_3049_label_13.png', 'id_3050_label_13.png', 'id_3051_label_14.png', 'id_3052_label_14.png', 'id_3053_label_15.png', 'id_3054_label_15.png', 'id_3055_label_16.png', 'id_3056_label_16.png', 'id_3057_label_17.png', 'id_3058_label_17.png', 'id_3059_label_18.png', 'id_3060_label_18.png', 'id_3061_label_19.png', 'id_3062_label_19.png', 'id_3063_label_20.png', 'id_3064_label_20.png', 'id_3065_label_21.png', 'id_3066_label_21.png', 'id_3067_label_22.png', 'id_3068_label_22.png', 'id_3069_label_23.png', 'id_3070_label_23.png', 'id_3071_label_24.png', 'id_3072_label_24.png', 'id_3073_label_25.png', 'id_3074_label_25.png', 'id_3075_label_26.png', 'id_3076_label_26.png', 'id_3077_label_27.png', 'id_3078_label_27.png', 'id_3079_label_28.png', 'id_3080_label_28.png', 'id_3081_label_1.png', 'id_3082_label_1.png', 'id_3083_label_2.png', 'id_3084_label_2.png', 'id_3085_label_3.png', 'id_3086_label_3.png', 'id_3087_label_4.png', 'id_3088_label_4.png', 'id_3089_label_5.png', 'id_3090_label_5.png', 'id_3091_label_6.png', 'id_3092_label_6.png', 'id_3093_label_7.png', 'id_3094_label_7.png', 'id_3095_label_8.png', 'id_3096_label_8.png', 'id_3097_label_9.png', 'id_3098_label_9.png', 'id_3099_label_10.png', 'id_3100_label_10.png', 'id_3101_label_11.png', 'id_3102_label_11.png', 'id_3103_label_12.png', 'id_3104_label_12.png', 'id_3105_label_13.png', 'id_3106_label_13.png', 'id_3107_label_14.png', 'id_3108_label_14.png', 'id_3109_label_15.png', 'id_3110_label_15.png', 'id_3111_label_16.png', 'id_3112_label_16.png', 'id_3113_label_17.png', 'id_3114_label_17.png', 'id_3115_label_18.png', 'id_3116_label_18.png', 'id_3117_label_19.png', 'id_3118_label_19.png', 'id_3119_label_20.png', 'id_3120_label_20.png', 'id_3121_label_21.png', 'id_3122_label_21.png', 'id_3123_label_22.png', 'id_3124_label_22.png', 'id_3125_label_23.png', 'id_3126_label_23.png', 'id_3127_label_24.png', 'id_3128_label_24.png', 'id_3129_label_25.png', 'id_3130_label_25.png', 'id_3131_label_26.png', 'id_3132_label_26.png', 'id_3133_label_27.png', 'id_3134_label_27.png', 'id_3135_label_28.png', 'id_3136_label_28.png', 'id_3137_label_1.png', 'id_3138_label_1.png', 'id_3139_label_2.png', 'id_3140_label_2.png', 'id_3141_label_3.png', 'id_3142_label_3.png', 'id_3143_label_4.png', 'id_3144_label_4.png', 'id_3145_label_5.png', 'id_3146_label_5.png', 'id_3147_label_6.png', 'id_3148_label_6.png', 'id_3149_label_7.png', 'id_3150_label_7.png', 'id_3151_label_8.png', 'id_3152_label_8.png', 'id_3153_label_9.png', 'id_3154_label_9.png', 'id_3155_label_10.png', 'id_3156_label_10.png', 'id_3157_label_11.png', 'id_3158_label_11.png', 'id_3159_label_12.png', 'id_3160_label_12.png', 'id_3161_label_13.png', 'id_3162_label_13.png', 'id_3163_label_14.png', 'id_3164_label_14.png', 'id_3165_label_15.png', 'id_3166_label_15.png', 'id_3167_label_16.png', 'id_3168_label_16.png', 'id_3169_label_17.png', 'id_3170_label_17.png', 'id_3171_label_18.png', 'id_3172_label_18.png', 'id_3173_label_19.png', 'id_3174_label_19.png', 'id_3175_label_20.png', 'id_3176_label_20.png', 'id_3177_label_21.png', 'id_3178_label_21.png', 'id_3179_label_22.png', 'id_3180_label_22.png', 'id_3181_label_23.png', 'id_3182_label_23.png', 'id_3183_label_24.png', 'id_3184_label_24.png', 'id_3185_label_25.png', 'id_3186_label_25.png', 'id_3187_label_26.png', 'id_3188_label_26.png', 'id_3189_label_27.png', 'id_3190_label_27.png', 'id_3191_label_28.png', 'id_3192_label_28.png', 'id_3193_label_1.png', 'id_3194_label_1.png', 'id_3195_label_2.png', 'id_3196_label_2.png', 'id_3197_label_3.png', 'id_3198_label_3.png', 'id_3199_label_4.png', 'id_3200_label_4.png', 'id_3201_label_5.png', 'id_3202_label_5.png', 'id_3203_label_6.png', 'id_3204_label_6.png', 'id_3205_label_7.png', 'id_3206_label_7.png', 'id_3207_label_8.png', 'id_3208_label_8.png', 'id_3209_label_9.png', 'id_3210_label_9.png', 'id_3211_label_10.png', 'id_3212_label_10.png', 'id_3213_label_11.png', 'id_3214_label_11.png', 'id_3215_label_12.png', 'id_3216_label_12.png', 'id_3217_label_13.png', 'id_3218_label_13.png', 'id_3219_label_14.png', 'id_3220_label_14.png', 'id_3221_label_15.png', 'id_3222_label_15.png', 'id_3223_label_16.png', 'id_3224_label_16.png', 'id_3225_label_17.png', 'id_3226_label_17.png', 'id_3227_label_18.png', 'id_3228_label_18.png', 'id_3229_label_19.png', 'id_3230_label_19.png', 'id_3231_label_20.png', 'id_3232_label_20.png', 'id_3233_label_21.png', 'id_3234_label_21.png', 'id_3235_label_22.png', 'id_3236_label_22.png', 'id_3237_label_23.png', 'id_3238_label_23.png', 'id_3239_label_24.png', 'id_3240_label_24.png', 'id_3241_label_25.png', 'id_3242_label_25.png', 'id_3243_label_26.png', 'id_3244_label_26.png', 'id_3245_label_27.png', 'id_3246_label_27.png', 'id_3247_label_28.png', 'id_3248_label_28.png', 'id_3249_label_1.png', 'id_3250_label_1.png', 'id_3251_label_2.png', 'id_3252_label_2.png', 'id_3253_label_3.png', 'id_3254_label_3.png', 'id_3255_label_4.png', 'id_3256_label_4.png', 'id_3257_label_5.png', 'id_3258_label_5.png', 'id_3259_label_6.png', 'id_3260_label_6.png', 'id_3261_label_7.png', 'id_3262_label_7.png', 'id_3263_label_8.png', 'id_3264_label_8.png', 'id_3265_label_9.png', 'id_3266_label_9.png', 'id_3267_label_10.png', 'id_3268_label_10.png', 'id_3269_label_11.png', 'id_3270_label_11.png', 'id_3271_label_12.png', 'id_3272_label_12.png', 'id_3273_label_13.png', 'id_3274_label_13.png', 'id_3275_label_14.png', 'id_3276_label_14.png', 'id_3277_label_15.png', 'id_3278_label_15.png', 'id_3279_label_16.png', 'id_3280_label_16.png', 'id_3281_label_17.png', 'id_3282_label_17.png', 'id_3283_label_18.png', 'id_3284_label_18.png', 'id_3285_label_19.png', 'id_3286_label_19.png', 'id_3287_label_20.png', 'id_3288_label_20.png', 'id_3289_label_21.png', 'id_3290_label_21.png', 'id_3291_label_22.png', 'id_3292_label_22.png', 'id_3293_label_23.png', 'id_3294_label_23.png', 'id_3295_label_24.png', 'id_3296_label_24.png', 'id_3297_label_25.png', 'id_3298_label_25.png', 'id_3299_label_26.png', 'id_3300_label_26.png', 'id_3301_label_27.png', 'id_3302_label_27.png', 'id_3303_label_28.png', 'id_3304_label_28.png', 'id_3305_label_1.png', 'id_3306_label_1.png', 'id_3307_label_2.png', 'id_3308_label_2.png', 'id_3309_label_3.png', 'id_3310_label_3.png', 'id_3311_label_4.png', 'id_3312_label_4.png', 'id_3313_label_5.png', 'id_3314_label_5.png', 'id_3315_label_6.png', 'id_3316_label_6.png', 'id_3317_label_7.png', 'id_3318_label_7.png', 'id_3319_label_8.png', 'id_3320_label_8.png', 'id_3321_label_9.png', 'id_3322_label_9.png', 'id_3323_label_10.png', 'id_3324_label_10.png', 'id_3325_label_11.png', 'id_3326_label_11.png', 'id_3327_label_12.png', 'id_3328_label_12.png', 'id_3329_label_13.png', 'id_3330_label_13.png', 'id_3331_label_14.png', 'id_3332_label_14.png', 'id_3333_label_15.png', 'id_3334_label_15.png', 'id_3335_label_16.png', 'id_3336_label_16.png', 'id_3337_label_17.png', 'id_3338_label_17.png', 'id_3339_label_18.png', 'id_3340_label_18.png', 'id_3341_label_19.png', 'id_3342_label_19.png', 'id_3343_label_20.png', 'id_3344_label_20.png', 'id_3345_label_21.png', 'id_3346_label_21.png', 'id_3347_label_22.png', 'id_3348_label_22.png', 'id_3349_label_23.png', 'id_3350_label_23.png', 'id_3351_label_24.png', 'id_3352_label_24.png', 'id_3353_label_25.png', 'id_3354_label_25.png', 'id_3355_label_26.png', 'id_3356_label_26.png', 'id_3357_label_27.png', 'id_3358_label_27.png', 'id_3359_label_28.png', 'id_3360_label_28.png']\n"
     ]
    }
   ],
   "source": [
    "print(LoTestImgNames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "LoTrueTestImgLabels = []\n",
    "for imgName in LoTestImgNames:\n",
    "    LoTrueTestImgLabels.append(imgName.split('_')[3].split('.')[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1'"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "LoTrueTestImgLabels[i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_cluster = pd.DataFrame(LoTestImgNames,columns=['image name'])\n",
    "image_cluster[\"clusterid\"] = clusters.labels_\n",
    "image_cluster[\"true label\"] = LoTrueTestImgLabels # 0 denotes cat and 1 denotes dog"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                image name  clusterid true label\n",
      "0         id_1_label_1.png          7          1\n",
      "1         id_2_label_1.png         24          1\n",
      "2         id_3_label_2.png          4          2\n",
      "3         id_4_label_2.png          4          2\n",
      "4         id_5_label_3.png         21          3\n",
      "...                    ...        ...        ...\n",
      "3355  id_3356_label_26.png          1         26\n",
      "3356  id_3357_label_27.png         17         27\n",
      "3357  id_3358_label_27.png         11         27\n",
      "3358  id_3359_label_28.png          9         28\n",
      "3359  id_3360_label_28.png         17         28\n",
      "\n",
      "[3360 rows x 3 columns]\n"
     ]
    }
   ],
   "source": [
    "print(image_cluster)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[7, 24, 24, 24, 24, 24, 24, 24, 24, 24, 2, 2, 24, 24, 24, 24, 7, 24, 24, 7, 24, 24, 7, 24, 2, 24, 7, 24, 24, 7, 24, 24, 24, 24, 24, 24, 2, 7, 24, 24, 24, 24, 24, 24, 24, 24, 7, 7, 24, 24, 7, 24, 3, 24, 24, 24, 24, 24, 24, 3, 24, 24, 10, 24, 24, 7, 7, 24, 24, 24, 7, 24, 24, 24, 7, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 7, 24, 24, 7, 24, 7, 7, 2, 2, 24, 7, 24, 24, 24, 7, 24, 24, 24, 7, 24, 24, 24, 24, 24, 24, 7, 24, 24, 24, 24, 24, 24, 24, 2, 2]\n"
     ]
    }
   ],
   "source": [
    "LoClusterIdsForLabel=[]\n",
    "for label in range(1,28):\n",
    "    clusterIdsFor_label = []\n",
    "    for i in range(len(LoTrueTestImgLabels)):\n",
    "        if int(LoTrueTestImgLabels[i]) == label:\n",
    "            clusterIdsFor_label.append(clusters.labels_[i])\n",
    "    LoClusterIdsForLabel.append(clusterIdsFor_label)\n",
    "\n",
    "print(LoClusterIdsForLabel[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "24"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max(set(clusterIdsFor_Alif), key=clusterIdsFor_Alif.count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "87"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clusterIdsFor_Alif.count(24)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "import collections\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Counter({7: 22, 24: 87, 2: 8, 3: 2, 10: 1})"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ctr = collections.Counter(clusterIdsFor_Alif)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.applications.inception_v3 import InceptionV3\n",
    "from tensorflow.keras.applications.inception_v3 import preprocess_input\n",
    "from tensorflow.keras.preprocessing import image\n",
    "from tensorflow.keras.preprocessing.image import img_to_array\n",
    "from sklearn.cluster import KMeans\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "import os\n",
    "import shutil\n",
    "from character_segmentation import segment\n",
    "from segmentation import line_horizontal_projection, word_vertical_projection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "recognizer.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "layer_outputs = [layer.output for layer in recognizer.layers]\n",
    "layer_outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_map_model = tf.keras.models.Model(inputs=recognizer.input, outputs=layer_outputs)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "recognizer.input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_maps = feature_map_model.predict(final_img)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_maps[8].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(feature_maps[8])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see if it cares about image size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_img.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "im_gray_th_otsu.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "type(im_gray_th_otsu)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sizeTestingimg = np.asarray(im_gray_th_otsu).reshape([-1, 364, 385, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_maps = feature_map_model.predict(sizeTestingimg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yes it does... so we have to change the shapes of each segment to fit this shape..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "testingmodel = InceptionV3(weights='imagenet', include_top=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "testingmodel.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "img=image.load_img(r\"D:\\School\\MedTech\\Sophomore Pre-Eng\\ISS296\\CNN\\images\\2alif.jpg\",target_size=(224,224))\n",
    "x = img_to_array(img)\n",
    "x=np.expand_dims(x,axis=0)\n",
    "x=preprocess_input(x)\n",
    "testingmodeloutput = testingmodel.predict(x).flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "testingmodeloutput.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(testingmodeloutput)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Raw Cell Format",
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}