DeepClustering_ResNet50.ipynb

{
  "metadata": {
    "kernelspec": {
      "language": "python",
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.7.12",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "colab": {
      "name": "DeepClustering_ResNet50.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "f0EdMqq-HVFI"
      ],
      "machine_shape": "hm"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "nbformat_minor": 0,
  "nbformat": 4,
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.models import Sequential, Model\n",
        "from tensorflow.keras import layers\n",
        "from keras.utils.vis_utils import plot_model\n",
        "from tensorflow.keras.models import Model,load_model\n",
        "from tensorflow.keras.layers import Flatten, Input, Dense, Conv2D, Conv2DTranspose,AveragePooling2D, Reshape, MaxPooling2D,UpSampling2D,LeakyReLU, concatenate, Dropout,BatchNormalization,Activation,ReLU\n",
        "from tensorflow.keras.callbacks import EarlyStopping,ModelCheckpoint\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.applications import ResNet50\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from sklearn.model_selection import train_test_split\n",
        "#OPTIMIZER\n",
        "from tensorflow.keras.optimizers import RMSprop,Adam,Optimizer,Optimizer, SGD\n",
        "#MODEL LAYERS\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.preprocessing.image import load_img, img_to_array\n",
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "import os\n",
        "import cv2 as cv\n",
        "import random\n",
        "\n",
        "from sklearn.decomposition import IncrementalPCA"
      ],
      "metadata": {
        "id": "knJqRFGbJ0xP",
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:42.890968Z",
          "iopub.execute_input": "2022-08-01T20:39:42.891328Z",
          "iopub.status.idle": "2022-08-01T20:39:42.901070Z",
          "shell.execute_reply.started": "2022-08-01T20:39:42.891298Z",
          "shell.execute_reply": "2022-08-01T20:39:42.899807Z"
        },
        "trusted": true
      },
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import cv2\n",
        "import matplotlib.pyplot as plt\n",
        "# Garbage Collector - use it like gc.collect()\n",
        "import gc\n",
        "import tensorflow as tf"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:44.427409Z",
          "iopub.execute_input": "2022-08-01T20:39:44.428006Z",
          "iopub.status.idle": "2022-08-01T20:39:44.433208Z",
          "shell.execute_reply.started": "2022-08-01T20:39:44.427970Z",
          "shell.execute_reply": "2022-08-01T20:39:44.432034Z"
        },
        "trusted": true,
        "id": "vFoi23DD6OEz"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "img_dim = 256\n",
        "def input_target_split():\n",
        "    train = []\n",
        "    images_list = os.listdir('/content/drive/MyDrive/agri_images')\n",
        "    for index in range(len(images_list)):\n",
        "        images_path = os.path.join('/content/drive/MyDrive/agri_images',images_list[index])\n",
        "        if(images_path == '/content/drive/MyDrive/agri_images/s2'):\n",
        "          break\n",
        "        image = load_img(os.path.join(images_path), target_size=(img_dim,img_dim)) \n",
        "        image = img_to_array(image)\n",
        "        image = image/255.0\n",
        "        train.append(image)\n",
        "    return np.array(train)"
      ],
      "metadata": {
        "id": "e9hgmBZJcDcU",
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:45.913522Z",
          "iopub.execute_input": "2022-08-01T20:39:45.914237Z",
          "iopub.status.idle": "2022-08-01T20:39:45.921754Z",
          "shell.execute_reply.started": "2022-08-01T20:39:45.914200Z",
          "shell.execute_reply": "2022-08-01T20:39:45.920438Z"
        },
        "trusted": true
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x = input_target_split()"
      ],
      "metadata": {
        "id": "NY36ylMtueyH",
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:48.822617Z",
          "iopub.execute_input": "2022-08-01T20:39:48.823071Z",
          "iopub.status.idle": "2022-08-01T20:40:10.561170Z",
          "shell.execute_reply.started": "2022-08-01T20:39:48.823031Z",
          "shell.execute_reply": "2022-08-01T20:40:10.555930Z"
        },
        "trusted": true
      },
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "datagen = ImageDataGenerator(rotation_range=90,width_shift_range=0.3,height_shift_range=0.3,zoom_range=[0.5, 1.5],horizontal_flip=True,vertical_flip=True)\n",
        "datagen.fit(x)"
      ],
      "metadata": {
        "id": "etOdSEe1upej",
        "execution": {
          "iopub.status.busy": "2022-08-01T20:40:10.569042Z",
          "iopub.execute_input": "2022-08-01T20:40:10.569507Z",
          "iopub.status.idle": "2022-08-01T20:40:11.847428Z",
          "shell.execute_reply.started": "2022-08-01T20:40:10.569452Z",
          "shell.execute_reply": "2022-08-01T20:40:11.846368Z"
        },
        "trusted": true
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "pretrained_model=ResNet50(weights=\"imagenet\", include_top=False, input_shape=(256,256,3))\n"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:40:11.848855Z",
          "iopub.execute_input": "2022-08-01T20:40:11.849541Z",
          "iopub.status.idle": "2022-08-01T20:40:12.758253Z",
          "shell.execute_reply.started": "2022-08-01T20:40:11.849475Z",
          "shell.execute_reply": "2022-08-01T20:40:12.757240Z"
        },
        "trusted": true,
        "id": "tbG2w4qX6OE3"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# pretrained_model.summary()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:18.712758Z",
          "iopub.status.idle": "2022-08-01T20:39:18.713552Z",
          "shell.execute_reply.started": "2022-08-01T20:39:18.713272Z",
          "shell.execute_reply": "2022-08-01T20:39:18.713297Z"
        },
        "trusted": true,
        "id": "t7aGeIi26OE4"
      },
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def CAE():\n",
        "    input_shape = (256, 256, 3)\n",
        "    last_layer = pretrained_model.output\n",
        "    set_trainable = False\n",
        "    for layer in pretrained_model.layers:\n",
        "        if layer.name in ['conv2d']:\n",
        "            set_trainable = True\n",
        "        if layer.name in ['max_pooling2d','max_pooling2d_1','average_pooling2d ','average_pooling2d_1','average_pooling2d_2','max_pooling2d_2','average_pooling2d_3','average_pooling2d_4','average_pooling2d_5','average_pooling2d_6','max_pooling2d_3','average_pooling2d_7','average_pooling2d_8',]:\n",
        "            layer.trainable = False\n",
        "    encoder_model = layers.Flatten()(last_layer)\n",
        "    encoder_model = layers.Dense(2048*2, kernel_initializer=\"glorot_uniform\",activation='relu')(encoder_model)\n",
        "    encoder_model = layers.Dense(100, kernel_initializer=\"glorot_uniform\", name='embedding')(encoder_model)  \n",
        "    \n",
        "    model_decoder = layers.Conv2DTranspose(512,(3,3),strides=(2, 2),padding='same')(last_layer)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder) \n",
        "    \n",
        "    model_decoder = layers.Conv2D(512,(3,3),strides=(1, 1),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder)\n",
        "    \n",
        "    model_decoder = layers.Conv2DTranspose(256,(3,3),strides=(2, 2),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder) \n",
        "    \n",
        "    model_decoder = layers.Conv2D(256,(3,3),strides=(1, 1),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder)\n",
        "    \n",
        "    model_decoder = layers.Conv2DTranspose(128,(3,3),strides=(2, 2),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder) \n",
        "    \n",
        "    model_decoder = layers.Conv2D(128,(3,3),strides=(1, 1),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder)\n",
        "    \n",
        "    model_decoder = layers.Conv2DTranspose(64,(3,3),strides=(2, 2),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder) \n",
        "    \n",
        "    model_decoder = layers.Conv2D(64,(3,3),strides=(1, 1),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder)\n",
        "    \n",
        "    model_decoder = layers.Conv2DTranspose(64,(3,3),strides=(2, 2),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder)\n",
        "    model_decoder = layers.Conv2D(3,(3,3),strides=(1, 1),padding='same')(model_decoder)\n",
        "    model_decoder = layers.LeakyReLU(0.1)(model_decoder)\n",
        "    model_decoder = layers.BatchNormalization()(model_decoder)\n",
        "    \n",
        "    model_decoder = Model(pretrained_model.input,model_decoder)\n",
        "    model_encoder = Model(pretrained_model.input,encoder_model)\n",
        "    model_encoder.summary()\n",
        "    model_decoder.summary()\n",
        "    return model_decoder , model_encoder"
      ],
      "metadata": {
        "id": "GhUVj41ruqQC",
        "execution": {
          "iopub.status.busy": "2022-08-01T20:40:12.761023Z",
          "iopub.execute_input": "2022-08-01T20:40:12.761401Z",
          "iopub.status.idle": "2022-08-01T20:40:12.782796Z",
          "shell.execute_reply.started": "2022-08-01T20:40:12.761363Z",
          "shell.execute_reply": "2022-08-01T20:40:12.781809Z"
        },
        "trusted": true
      },
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\"\"\"import tensorflow as tf\n",
        "autoencoder.load_weights(\"./model_5.h5\")\"\"\""
      ],
      "metadata": {
        "id": "CiPKmEbS1gvV",
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:18.717192Z",
          "iopub.status.idle": "2022-08-01T20:39:18.717912Z",
          "shell.execute_reply.started": "2022-08-01T20:39:18.717650Z",
          "shell.execute_reply": "2022-08-01T20:39:18.717675Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "outputId": "efe2aa53-c6e8-4457-e78e-0de9473c1445"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'import tensorflow as tf\\nautoencoder.load_weights(\"./model_5.h5\")'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "gc.collect()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:40:12.784270Z",
          "iopub.execute_input": "2022-08-01T20:40:12.784823Z",
          "iopub.status.idle": "2022-08-01T20:40:13.042974Z",
          "shell.execute_reply.started": "2022-08-01T20:40:12.784786Z",
          "shell.execute_reply": "2022-08-01T20:40:13.041928Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MOk_XEAH6OE6",
        "outputId": "bc4988a5-b442-4595-a946-4342407b18a6"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "68124"
            ]
          },
          "metadata": {},
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder,encoder = CAE()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:40:23.166746Z",
          "iopub.execute_input": "2022-08-01T20:40:23.167662Z",
          "iopub.status.idle": "2022-08-01T20:40:23.402568Z",
          "shell.execute_reply.started": "2022-08-01T20:40:23.167612Z",
          "shell.execute_reply": "2022-08-01T20:40:23.401556Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4gAemgsW6OE6",
        "outputId": "4522f841-1ce4-41d4-f9b5-cd00c3aea9f9"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_4\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_2 (InputLayer)           [(None, 256, 256, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " conv1_pad (ZeroPadding2D)      (None, 262, 262, 3)  0           ['input_2[0][0]']                \n",
            "                                                                                                  \n",
            " conv1_conv (Conv2D)            (None, 128, 128, 64  9472        ['conv1_pad[0][0]']              \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv1_bn (BatchNormalization)  (None, 128, 128, 64  256         ['conv1_conv[0][0]']             \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv1_relu (Activation)        (None, 128, 128, 64  0           ['conv1_bn[0][0]']               \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " pool1_pad (ZeroPadding2D)      (None, 130, 130, 64  0           ['conv1_relu[0][0]']             \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " pool1_pool (MaxPooling2D)      (None, 64, 64, 64)   0           ['pool1_pad[0][0]']              \n",
            "                                                                                                  \n",
            " conv2_block1_1_conv (Conv2D)   (None, 64, 64, 64)   4160        ['pool1_pool[0][0]']             \n",
            "                                                                                                  \n",
            " conv2_block1_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block1_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_0_conv (Conv2D)   (None, 64, 64, 256)  16640       ['pool1_pool[0][0]']             \n",
            "                                                                                                  \n",
            " conv2_block1_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block1_0_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_add (Add)         (None, 64, 64, 256)  0           ['conv2_block1_0_bn[0][0]',      \n",
            "                                                                  'conv2_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block1_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block2_1_conv (Conv2D)   (None, 64, 64, 64)   16448       ['conv2_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block2_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block2_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block2_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_add (Add)         (None, 64, 64, 256)  0           ['conv2_block1_out[0][0]',       \n",
            "                                                                  'conv2_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block2_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block3_1_conv (Conv2D)   (None, 64, 64, 64)   16448       ['conv2_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block3_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block3_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block3_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_add (Add)         (None, 64, 64, 256)  0           ['conv2_block2_out[0][0]',       \n",
            "                                                                  'conv2_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block3_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_1_conv (Conv2D)   (None, 32, 32, 128)  32896       ['conv2_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block1_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_0_conv (Conv2D)   (None, 32, 32, 512)  131584      ['conv2_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block1_0_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_add (Add)         (None, 32, 32, 512)  0           ['conv3_block1_0_bn[0][0]',      \n",
            "                                                                  'conv3_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block1_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block2_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block2_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block2_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block2_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_add (Add)         (None, 32, 32, 512)  0           ['conv3_block1_out[0][0]',       \n",
            "                                                                  'conv3_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block2_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block3_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block3_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block3_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block3_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_add (Add)         (None, 32, 32, 512)  0           ['conv3_block2_out[0][0]',       \n",
            "                                                                  'conv3_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block3_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block4_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block4_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block4_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block4_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block4_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block4_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block4_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block4_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block4_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block4_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block4_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_add (Add)         (None, 32, 32, 512)  0           ['conv3_block3_out[0][0]',       \n",
            "                                                                  'conv3_block4_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block4_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block4_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block1_1_conv (Conv2D)   (None, 16, 16, 256)  131328      ['conv3_block4_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block1_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block1_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block1_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block1_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_0_conv (Conv2D)   (None, 16, 16, 1024  525312      ['conv3_block4_out[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block1_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_0_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block1_0_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block1_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_add (Add)         (None, 16, 16, 1024  0           ['conv4_block1_0_bn[0][0]',      \n",
            "                                )                                 'conv4_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block1_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block1_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block2_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block2_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block2_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block2_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block2_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block2_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_add (Add)         (None, 16, 16, 1024  0           ['conv4_block1_out[0][0]',       \n",
            "                                )                                 'conv4_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block2_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block2_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block3_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block3_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block3_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block3_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block3_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block3_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_add (Add)         (None, 16, 16, 1024  0           ['conv4_block2_out[0][0]',       \n",
            "                                )                                 'conv4_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block3_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block3_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block4_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block4_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block4_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block4_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block4_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block4_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block4_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block4_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block4_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block4_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block4_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_add (Add)         (None, 16, 16, 1024  0           ['conv4_block3_out[0][0]',       \n",
            "                                )                                 'conv4_block4_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block4_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block4_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block4_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block5_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block5_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block5_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block5_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block5_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block5_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block5_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block5_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block5_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block5_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block5_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_add (Add)         (None, 16, 16, 1024  0           ['conv4_block4_out[0][0]',       \n",
            "                                )                                 'conv4_block5_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block5_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block5_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block5_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block6_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block6_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block6_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block6_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block6_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block6_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block6_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block6_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block6_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block6_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block6_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_add (Add)         (None, 16, 16, 1024  0           ['conv4_block5_out[0][0]',       \n",
            "                                )                                 'conv4_block6_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block6_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block6_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv5_block1_1_conv (Conv2D)   (None, 8, 8, 512)    524800      ['conv4_block6_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block1_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block1_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_0_conv (Conv2D)   (None, 8, 8, 2048)   2099200     ['conv4_block6_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block1_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block1_0_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block1_0_bn[0][0]',      \n",
            "                                                                  'conv5_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block1_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block2_1_conv (Conv2D)   (None, 8, 8, 512)    1049088     ['conv5_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block2_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block2_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block2_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block1_out[0][0]',       \n",
            "                                                                  'conv5_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block2_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block3_1_conv (Conv2D)   (None, 8, 8, 512)    1049088     ['conv5_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block3_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block3_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block3_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block2_out[0][0]',       \n",
            "                                                                  'conv5_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block3_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " flatten_1 (Flatten)            (None, 131072)       0           ['conv5_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " dense_1 (Dense)                (None, 4096)         536875008   ['flatten_1[0][0]']              \n",
            "                                                                                                  \n",
            " embedding (Dense)              (None, 100)          409700      ['dense_1[0][0]']                \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 560,872,420\n",
            "Trainable params: 560,819,300\n",
            "Non-trainable params: 53,120\n",
            "__________________________________________________________________________________________________\n",
            "Model: \"model_3\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_2 (InputLayer)           [(None, 256, 256, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " conv1_pad (ZeroPadding2D)      (None, 262, 262, 3)  0           ['input_2[0][0]']                \n",
            "                                                                                                  \n",
            " conv1_conv (Conv2D)            (None, 128, 128, 64  9472        ['conv1_pad[0][0]']              \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv1_bn (BatchNormalization)  (None, 128, 128, 64  256         ['conv1_conv[0][0]']             \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv1_relu (Activation)        (None, 128, 128, 64  0           ['conv1_bn[0][0]']               \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " pool1_pad (ZeroPadding2D)      (None, 130, 130, 64  0           ['conv1_relu[0][0]']             \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " pool1_pool (MaxPooling2D)      (None, 64, 64, 64)   0           ['pool1_pad[0][0]']              \n",
            "                                                                                                  \n",
            " conv2_block1_1_conv (Conv2D)   (None, 64, 64, 64)   4160        ['pool1_pool[0][0]']             \n",
            "                                                                                                  \n",
            " conv2_block1_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block1_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_0_conv (Conv2D)   (None, 64, 64, 256)  16640       ['pool1_pool[0][0]']             \n",
            "                                                                                                  \n",
            " conv2_block1_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block1_0_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_add (Add)         (None, 64, 64, 256)  0           ['conv2_block1_0_bn[0][0]',      \n",
            "                                                                  'conv2_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block1_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block2_1_conv (Conv2D)   (None, 64, 64, 64)   16448       ['conv2_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block2_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block2_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block2_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_add (Add)         (None, 64, 64, 256)  0           ['conv2_block1_out[0][0]',       \n",
            "                                                                  'conv2_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block2_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block3_1_conv (Conv2D)   (None, 64, 64, 64)   16448       ['conv2_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block3_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block3_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block3_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_add (Add)         (None, 64, 64, 256)  0           ['conv2_block2_out[0][0]',       \n",
            "                                                                  'conv2_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block3_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_1_conv (Conv2D)   (None, 32, 32, 128)  32896       ['conv2_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block1_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_0_conv (Conv2D)   (None, 32, 32, 512)  131584      ['conv2_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block1_0_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_add (Add)         (None, 32, 32, 512)  0           ['conv3_block1_0_bn[0][0]',      \n",
            "                                                                  'conv3_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block1_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block2_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block2_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block2_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block2_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_add (Add)         (None, 32, 32, 512)  0           ['conv3_block1_out[0][0]',       \n",
            "                                                                  'conv3_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block2_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block3_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block3_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block3_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block3_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_add (Add)         (None, 32, 32, 512)  0           ['conv3_block2_out[0][0]',       \n",
            "                                                                  'conv3_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block3_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block4_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block4_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block4_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block4_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block4_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block4_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block4_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block4_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block4_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block4_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block4_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_add (Add)         (None, 32, 32, 512)  0           ['conv3_block3_out[0][0]',       \n",
            "                                                                  'conv3_block4_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block4_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block4_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block1_1_conv (Conv2D)   (None, 16, 16, 256)  131328      ['conv3_block4_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block1_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block1_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block1_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block1_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_0_conv (Conv2D)   (None, 16, 16, 1024  525312      ['conv3_block4_out[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block1_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_0_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block1_0_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block1_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_add (Add)         (None, 16, 16, 1024  0           ['conv4_block1_0_bn[0][0]',      \n",
            "                                )                                 'conv4_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block1_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block1_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block2_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block2_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block2_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block2_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block2_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block2_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_add (Add)         (None, 16, 16, 1024  0           ['conv4_block1_out[0][0]',       \n",
            "                                )                                 'conv4_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block2_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block2_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block3_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block3_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block3_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block3_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block3_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block3_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_add (Add)         (None, 16, 16, 1024  0           ['conv4_block2_out[0][0]',       \n",
            "                                )                                 'conv4_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block3_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block3_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block4_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block4_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block4_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block4_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block4_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block4_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block4_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block4_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block4_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block4_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block4_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_add (Add)         (None, 16, 16, 1024  0           ['conv4_block3_out[0][0]',       \n",
            "                                )                                 'conv4_block4_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block4_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block4_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block4_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block5_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block5_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block5_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block5_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block5_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block5_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block5_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block5_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block5_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block5_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block5_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_add (Add)         (None, 16, 16, 1024  0           ['conv4_block4_out[0][0]',       \n",
            "                                )                                 'conv4_block5_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block5_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block5_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block5_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block6_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block6_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block6_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block6_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block6_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block6_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block6_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block6_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block6_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block6_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block6_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_add (Add)         (None, 16, 16, 1024  0           ['conv4_block5_out[0][0]',       \n",
            "                                )                                 'conv4_block6_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block6_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block6_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv5_block1_1_conv (Conv2D)   (None, 8, 8, 512)    524800      ['conv4_block6_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block1_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block1_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_0_conv (Conv2D)   (None, 8, 8, 2048)   2099200     ['conv4_block6_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block1_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block1_0_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block1_0_bn[0][0]',      \n",
            "                                                                  'conv5_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block1_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block2_1_conv (Conv2D)   (None, 8, 8, 512)    1049088     ['conv5_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block2_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block2_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block2_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block1_out[0][0]',       \n",
            "                                                                  'conv5_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block2_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block3_1_conv (Conv2D)   (None, 8, 8, 512)    1049088     ['conv5_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block3_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block3_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block3_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block2_out[0][0]',       \n",
            "                                                                  'conv5_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block3_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv2d_transpose_5 (Conv2DTran  (None, 16, 16, 512)  9437696    ['conv5_block3_out[0][0]']       \n",
            " spose)                                                                                           \n",
            "                                                                                                  \n",
            " leaky_re_lu_10 (LeakyReLU)     (None, 16, 16, 512)  0           ['conv2d_transpose_5[0][0]']     \n",
            "                                                                                                  \n",
            " batch_normalization_10 (BatchN  (None, 16, 16, 512)  2048       ['leaky_re_lu_10[0][0]']         \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv2d_5 (Conv2D)              (None, 16, 16, 512)  2359808     ['batch_normalization_10[0][0]'] \n",
            "                                                                                                  \n",
            " leaky_re_lu_11 (LeakyReLU)     (None, 16, 16, 512)  0           ['conv2d_5[0][0]']               \n",
            "                                                                                                  \n",
            " batch_normalization_11 (BatchN  (None, 16, 16, 512)  2048       ['leaky_re_lu_11[0][0]']         \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv2d_transpose_6 (Conv2DTran  (None, 32, 32, 256)  1179904    ['batch_normalization_11[0][0]'] \n",
            " spose)                                                                                           \n",
            "                                                                                                  \n",
            " leaky_re_lu_12 (LeakyReLU)     (None, 32, 32, 256)  0           ['conv2d_transpose_6[0][0]']     \n",
            "                                                                                                  \n",
            " batch_normalization_12 (BatchN  (None, 32, 32, 256)  1024       ['leaky_re_lu_12[0][0]']         \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv2d_6 (Conv2D)              (None, 32, 32, 256)  590080      ['batch_normalization_12[0][0]'] \n",
            "                                                                                                  \n",
            " leaky_re_lu_13 (LeakyReLU)     (None, 32, 32, 256)  0           ['conv2d_6[0][0]']               \n",
            "                                                                                                  \n",
            " batch_normalization_13 (BatchN  (None, 32, 32, 256)  1024       ['leaky_re_lu_13[0][0]']         \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv2d_transpose_7 (Conv2DTran  (None, 64, 64, 128)  295040     ['batch_normalization_13[0][0]'] \n",
            " spose)                                                                                           \n",
            "                                                                                                  \n",
            " leaky_re_lu_14 (LeakyReLU)     (None, 64, 64, 128)  0           ['conv2d_transpose_7[0][0]']     \n",
            "                                                                                                  \n",
            " batch_normalization_14 (BatchN  (None, 64, 64, 128)  512        ['leaky_re_lu_14[0][0]']         \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv2d_7 (Conv2D)              (None, 64, 64, 128)  147584      ['batch_normalization_14[0][0]'] \n",
            "                                                                                                  \n",
            " leaky_re_lu_15 (LeakyReLU)     (None, 64, 64, 128)  0           ['conv2d_7[0][0]']               \n",
            "                                                                                                  \n",
            " batch_normalization_15 (BatchN  (None, 64, 64, 128)  512        ['leaky_re_lu_15[0][0]']         \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv2d_transpose_8 (Conv2DTran  (None, 128, 128, 64  73792      ['batch_normalization_15[0][0]'] \n",
            " spose)                         )                                                                 \n",
            "                                                                                                  \n",
            " leaky_re_lu_16 (LeakyReLU)     (None, 128, 128, 64  0           ['conv2d_transpose_8[0][0]']     \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " batch_normalization_16 (BatchN  (None, 128, 128, 64  256        ['leaky_re_lu_16[0][0]']         \n",
            " ormalization)                  )                                                                 \n",
            "                                                                                                  \n",
            " conv2d_8 (Conv2D)              (None, 128, 128, 64  36928       ['batch_normalization_16[0][0]'] \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " leaky_re_lu_17 (LeakyReLU)     (None, 128, 128, 64  0           ['conv2d_8[0][0]']               \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " batch_normalization_17 (BatchN  (None, 128, 128, 64  256        ['leaky_re_lu_17[0][0]']         \n",
            " ormalization)                  )                                                                 \n",
            "                                                                                                  \n",
            " conv2d_transpose_9 (Conv2DTran  (None, 256, 256, 64  36928      ['batch_normalization_17[0][0]'] \n",
            " spose)                         )                                                                 \n",
            "                                                                                                  \n",
            " leaky_re_lu_18 (LeakyReLU)     (None, 256, 256, 64  0           ['conv2d_transpose_9[0][0]']     \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " batch_normalization_18 (BatchN  (None, 256, 256, 64  256        ['leaky_re_lu_18[0][0]']         \n",
            " ormalization)                  )                                                                 \n",
            "                                                                                                  \n",
            " conv2d_9 (Conv2D)              (None, 256, 256, 3)  1731        ['batch_normalization_18[0][0]'] \n",
            "                                                                                                  \n",
            " leaky_re_lu_19 (LeakyReLU)     (None, 256, 256, 3)  0           ['conv2d_9[0][0]']               \n",
            "                                                                                                  \n",
            " batch_normalization_19 (BatchN  (None, 256, 256, 3)  12         ['leaky_re_lu_19[0][0]']         \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 37,755,151\n",
            "Trainable params: 37,698,057\n",
            "Non-trainable params: 57,094\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder.compile(optimizer = Adam(learning_rate=0.001), loss = ['mse'], metrics=['accuracy'])"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:40:25.640891Z",
          "iopub.execute_input": "2022-08-01T20:40:25.641263Z",
          "iopub.status.idle": "2022-08-01T20:40:25.660324Z",
          "shell.execute_reply.started": "2022-08-01T20:40:25.641232Z",
          "shell.execute_reply": "2022-08-01T20:40:25.659224Z"
        },
        "trusted": true,
        "id": "folPjkef6OE6"
      },
      "execution_count": 58,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "gc.collect()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1nbJSFfd_-ye",
        "outputId": "bec91965-2503-44ed-e702-47e72e9c1b3e"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "8381"
            ]
          },
          "metadata": {},
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder.fit_generator(datagen.flow(x,x,batch_size=16),epochs=200, steps_per_epoch=20)\n",
        "autoencoder.save_weights('model_auto.h5')"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:40:35.195490Z",
          "iopub.execute_input": "2022-08-01T20:40:35.196125Z",
          "iopub.status.idle": "2022-08-01T20:55:16.006229Z",
          "shell.execute_reply.started": "2022-08-01T20:40:35.196089Z",
          "shell.execute_reply": "2022-08-01T20:55:16.005173Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TJ7xKEBV6OE7",
        "outputId": "5e7e1f15-6c50-43ce-8775-0eed2769de41"
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:1: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/200\n",
            "20/20 [==============================] - 10s 232ms/step - loss: 1.0227 - accuracy: 0.3006\n",
            "Epoch 2/200\n",
            "20/20 [==============================] - 6s 272ms/step - loss: 0.9533 - accuracy: 0.2196\n",
            "Epoch 3/200\n",
            "20/20 [==============================] - 5s 246ms/step - loss: 0.9065 - accuracy: 0.2747\n",
            "Epoch 4/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.8617 - accuracy: 0.3285\n",
            "Epoch 5/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.8130 - accuracy: 0.3967\n",
            "Epoch 6/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.7717 - accuracy: 0.4158\n",
            "Epoch 7/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.7396 - accuracy: 0.3665\n",
            "Epoch 8/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.7047 - accuracy: 0.4357\n",
            "Epoch 9/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.6693 - accuracy: 0.3819\n",
            "Epoch 10/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.6247 - accuracy: 0.3911\n",
            "Epoch 11/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.6015 - accuracy: 0.4319\n",
            "Epoch 12/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.5567 - accuracy: 0.4574\n",
            "Epoch 13/200\n",
            "20/20 [==============================] - 5s 235ms/step - loss: 0.5369 - accuracy: 0.4153\n",
            "Epoch 14/200\n",
            "20/20 [==============================] - 5s 246ms/step - loss: 0.5141 - accuracy: 0.3897\n",
            "Epoch 15/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.4896 - accuracy: 0.4266\n",
            "Epoch 16/200\n",
            "20/20 [==============================] - 5s 238ms/step - loss: 0.4576 - accuracy: 0.4541\n",
            "Epoch 17/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.4394 - accuracy: 0.4619\n",
            "Epoch 18/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.4125 - accuracy: 0.4462\n",
            "Epoch 19/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.3895 - accuracy: 0.4886\n",
            "Epoch 20/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.3766 - accuracy: 0.4123\n",
            "Epoch 21/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.3554 - accuracy: 0.4320\n",
            "Epoch 22/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.3337 - accuracy: 0.4796\n",
            "Epoch 23/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.3168 - accuracy: 0.4510\n",
            "Epoch 24/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.3020 - accuracy: 0.4322\n",
            "Epoch 25/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.2891 - accuracy: 0.4618\n",
            "Epoch 26/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.2724 - accuracy: 0.4756\n",
            "Epoch 27/200\n",
            "20/20 [==============================] - 5s 230ms/step - loss: 0.2580 - accuracy: 0.4320\n",
            "Epoch 28/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.2395 - accuracy: 0.5003\n",
            "Epoch 29/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.2254 - accuracy: 0.4692\n",
            "Epoch 30/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.2162 - accuracy: 0.5334\n",
            "Epoch 31/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.2016 - accuracy: 0.4859\n",
            "Epoch 32/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.1931 - accuracy: 0.4619\n",
            "Epoch 33/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.1861 - accuracy: 0.3644\n",
            "Epoch 34/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.1712 - accuracy: 0.4439\n",
            "Epoch 35/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.1645 - accuracy: 0.4532\n",
            "Epoch 36/200\n",
            "20/20 [==============================] - 5s 246ms/step - loss: 0.1527 - accuracy: 0.5013\n",
            "Epoch 37/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.1438 - accuracy: 0.4724\n",
            "Epoch 38/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.1337 - accuracy: 0.4850\n",
            "Epoch 39/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.1251 - accuracy: 0.5425\n",
            "Epoch 40/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.1161 - accuracy: 0.5087\n",
            "Epoch 41/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.1120 - accuracy: 0.5240\n",
            "Epoch 42/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.1032 - accuracy: 0.5155\n",
            "Epoch 43/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.1003 - accuracy: 0.4906\n",
            "Epoch 44/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.0898 - accuracy: 0.5332\n",
            "Epoch 45/200\n",
            "20/20 [==============================] - 5s 237ms/step - loss: 0.0862 - accuracy: 0.5478\n",
            "Epoch 46/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0792 - accuracy: 0.5424\n",
            "Epoch 47/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0756 - accuracy: 0.5525\n",
            "Epoch 48/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0717 - accuracy: 0.5691\n",
            "Epoch 49/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0655 - accuracy: 0.5799\n",
            "Epoch 50/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0630 - accuracy: 0.5676\n",
            "Epoch 51/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0591 - accuracy: 0.5555\n",
            "Epoch 52/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0538 - accuracy: 0.5759\n",
            "Epoch 53/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0503 - accuracy: 0.5803\n",
            "Epoch 54/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0469 - accuracy: 0.5788\n",
            "Epoch 55/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0444 - accuracy: 0.5778\n",
            "Epoch 56/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0412 - accuracy: 0.5817\n",
            "Epoch 57/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0405 - accuracy: 0.5743\n",
            "Epoch 58/200\n",
            "20/20 [==============================] - 5s 239ms/step - loss: 0.0346 - accuracy: 0.6026\n",
            "Epoch 59/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0348 - accuracy: 0.6231\n",
            "Epoch 60/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0324 - accuracy: 0.5886\n",
            "Epoch 61/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0299 - accuracy: 0.5915\n",
            "Epoch 62/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0278 - accuracy: 0.6366\n",
            "Epoch 63/200\n",
            "20/20 [==============================] - 5s 239ms/step - loss: 0.0267 - accuracy: 0.6049\n",
            "Epoch 64/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0240 - accuracy: 0.6259\n",
            "Epoch 65/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0222 - accuracy: 0.6461\n",
            "Epoch 66/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0235 - accuracy: 0.6173\n",
            "Epoch 67/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0224 - accuracy: 0.6122\n",
            "Epoch 68/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0196 - accuracy: 0.6431\n",
            "Epoch 69/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0185 - accuracy: 0.6173\n",
            "Epoch 70/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0179 - accuracy: 0.6478\n",
            "Epoch 71/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0176 - accuracy: 0.6270\n",
            "Epoch 72/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0163 - accuracy: 0.6338\n",
            "Epoch 73/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0147 - accuracy: 0.6496\n",
            "Epoch 74/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0153 - accuracy: 0.6291\n",
            "Epoch 75/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0158 - accuracy: 0.6671\n",
            "Epoch 76/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0140 - accuracy: 0.6891\n",
            "Epoch 77/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0129 - accuracy: 0.6714\n",
            "Epoch 78/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0128 - accuracy: 0.6788\n",
            "Epoch 79/200\n",
            "20/20 [==============================] - 5s 246ms/step - loss: 0.0124 - accuracy: 0.6750\n",
            "Epoch 80/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0128 - accuracy: 0.6798\n",
            "Epoch 81/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0115 - accuracy: 0.6964\n",
            "Epoch 82/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0117 - accuracy: 0.6876\n",
            "Epoch 83/200\n",
            "20/20 [==============================] - 5s 235ms/step - loss: 0.0116 - accuracy: 0.6776\n",
            "Epoch 84/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0106 - accuracy: 0.6895\n",
            "Epoch 85/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0110 - accuracy: 0.6630\n",
            "Epoch 86/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0109 - accuracy: 0.6970\n",
            "Epoch 87/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0104 - accuracy: 0.6834\n",
            "Epoch 88/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0105 - accuracy: 0.6919\n",
            "Epoch 89/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0098 - accuracy: 0.6816\n",
            "Epoch 90/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0098 - accuracy: 0.7042\n",
            "Epoch 91/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0098 - accuracy: 0.6776\n",
            "Epoch 92/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0099 - accuracy: 0.6895\n",
            "Epoch 93/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0091 - accuracy: 0.7216\n",
            "Epoch 94/200\n",
            "20/20 [==============================] - 5s 236ms/step - loss: 0.0096 - accuracy: 0.6885\n",
            "Epoch 95/200\n",
            "20/20 [==============================] - 5s 235ms/step - loss: 0.0091 - accuracy: 0.7014\n",
            "Epoch 96/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0099 - accuracy: 0.6856\n",
            "Epoch 97/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0093 - accuracy: 0.6960\n",
            "Epoch 98/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0098 - accuracy: 0.6979\n",
            "Epoch 99/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0099 - accuracy: 0.6728\n",
            "Epoch 100/200\n",
            "20/20 [==============================] - 5s 246ms/step - loss: 0.0087 - accuracy: 0.6980\n",
            "Epoch 101/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0090 - accuracy: 0.7036\n",
            "Epoch 102/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0095 - accuracy: 0.6912\n",
            "Epoch 103/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0093 - accuracy: 0.6964\n",
            "Epoch 104/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0090 - accuracy: 0.7025\n",
            "Epoch 105/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0092 - accuracy: 0.6873\n",
            "Epoch 106/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0092 - accuracy: 0.6869\n",
            "Epoch 107/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0092 - accuracy: 0.6830\n",
            "Epoch 108/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0097 - accuracy: 0.6913\n",
            "Epoch 109/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0091 - accuracy: 0.7059\n",
            "Epoch 110/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0099 - accuracy: 0.6928\n",
            "Epoch 111/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0089 - accuracy: 0.6894\n",
            "Epoch 112/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0096 - accuracy: 0.6939\n",
            "Epoch 113/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0094 - accuracy: 0.6656\n",
            "Epoch 114/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0090 - accuracy: 0.7023\n",
            "Epoch 115/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0091 - accuracy: 0.6899\n",
            "Epoch 116/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0089 - accuracy: 0.6954\n",
            "Epoch 117/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0090 - accuracy: 0.6886\n",
            "Epoch 118/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.0089 - accuracy: 0.7082\n",
            "Epoch 119/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0087 - accuracy: 0.7089\n",
            "Epoch 120/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0092 - accuracy: 0.6933\n",
            "Epoch 121/200\n",
            "20/20 [==============================] - 5s 237ms/step - loss: 0.0093 - accuracy: 0.6771\n",
            "Epoch 122/200\n",
            "20/20 [==============================] - 5s 236ms/step - loss: 0.0088 - accuracy: 0.7001\n",
            "Epoch 123/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0090 - accuracy: 0.6928\n",
            "Epoch 124/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0092 - accuracy: 0.7103\n",
            "Epoch 125/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0093 - accuracy: 0.6963\n",
            "Epoch 126/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0091 - accuracy: 0.6821\n",
            "Epoch 127/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0093 - accuracy: 0.6970\n",
            "Epoch 128/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0093 - accuracy: 0.7033\n",
            "Epoch 129/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0093 - accuracy: 0.6815\n",
            "Epoch 130/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0090 - accuracy: 0.7014\n",
            "Epoch 131/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0092 - accuracy: 0.7043\n",
            "Epoch 132/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0091 - accuracy: 0.6905\n",
            "Epoch 133/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0094 - accuracy: 0.6893\n",
            "Epoch 134/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0088 - accuracy: 0.6825\n",
            "Epoch 135/200\n",
            "20/20 [==============================] - 5s 230ms/step - loss: 0.0094 - accuracy: 0.6998\n",
            "Epoch 136/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0088 - accuracy: 0.6979\n",
            "Epoch 137/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0088 - accuracy: 0.7075\n",
            "Epoch 138/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0086 - accuracy: 0.7078\n",
            "Epoch 139/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0092 - accuracy: 0.6884\n",
            "Epoch 140/200\n",
            "20/20 [==============================] - 5s 235ms/step - loss: 0.0092 - accuracy: 0.7046\n",
            "Epoch 141/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0089 - accuracy: 0.6868\n",
            "Epoch 142/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.0093 - accuracy: 0.6902\n",
            "Epoch 143/200\n",
            "20/20 [==============================] - 5s 242ms/step - loss: 0.0085 - accuracy: 0.6995\n",
            "Epoch 144/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0092 - accuracy: 0.7074\n",
            "Epoch 145/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0089 - accuracy: 0.7088\n",
            "Epoch 146/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0091 - accuracy: 0.6812\n",
            "Epoch 147/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0095 - accuracy: 0.6757\n",
            "Epoch 148/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0087 - accuracy: 0.6991\n",
            "Epoch 149/200\n",
            "20/20 [==============================] - 5s 238ms/step - loss: 0.0090 - accuracy: 0.6983\n",
            "Epoch 150/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0097 - accuracy: 0.6927\n",
            "Epoch 151/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0096 - accuracy: 0.6934\n",
            "Epoch 152/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0091 - accuracy: 0.7196\n",
            "Epoch 153/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.0088 - accuracy: 0.7000\n",
            "Epoch 154/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0092 - accuracy: 0.6872\n",
            "Epoch 155/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0088 - accuracy: 0.6917\n",
            "Epoch 156/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0093 - accuracy: 0.6780\n",
            "Epoch 157/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0090 - accuracy: 0.6975\n",
            "Epoch 158/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0096 - accuracy: 0.7240\n",
            "Epoch 159/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0093 - accuracy: 0.6881\n",
            "Epoch 160/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0095 - accuracy: 0.7033\n",
            "Epoch 161/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0095 - accuracy: 0.6922\n",
            "Epoch 162/200\n",
            "20/20 [==============================] - 5s 236ms/step - loss: 0.0088 - accuracy: 0.6881\n",
            "Epoch 163/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0088 - accuracy: 0.6789\n",
            "Epoch 164/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0092 - accuracy: 0.7001\n",
            "Epoch 165/200\n",
            "20/20 [==============================] - 5s 243ms/step - loss: 0.0092 - accuracy: 0.6941\n",
            "Epoch 166/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0092 - accuracy: 0.7044\n",
            "Epoch 167/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0088 - accuracy: 0.6845\n",
            "Epoch 168/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0092 - accuracy: 0.6882\n",
            "Epoch 169/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0089 - accuracy: 0.7284\n",
            "Epoch 170/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0099 - accuracy: 0.6721\n",
            "Epoch 171/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0090 - accuracy: 0.6890\n",
            "Epoch 172/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0093 - accuracy: 0.6953\n",
            "Epoch 173/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0097 - accuracy: 0.6950\n",
            "Epoch 174/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0094 - accuracy: 0.6880\n",
            "Epoch 175/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0088 - accuracy: 0.6888\n",
            "Epoch 176/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0090 - accuracy: 0.6786\n",
            "Epoch 177/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0090 - accuracy: 0.6915\n",
            "Epoch 178/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0096 - accuracy: 0.6832\n",
            "Epoch 179/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0088 - accuracy: 0.6838\n",
            "Epoch 180/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0090 - accuracy: 0.6741\n",
            "Epoch 181/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0097 - accuracy: 0.6876\n",
            "Epoch 182/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.0089 - accuracy: 0.7113\n",
            "Epoch 183/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0090 - accuracy: 0.7071\n",
            "Epoch 184/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0094 - accuracy: 0.7010\n",
            "Epoch 185/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0090 - accuracy: 0.6913\n",
            "Epoch 186/200\n",
            "20/20 [==============================] - 5s 237ms/step - loss: 0.0092 - accuracy: 0.7080\n",
            "Epoch 187/200\n",
            "20/20 [==============================] - 5s 231ms/step - loss: 0.0093 - accuracy: 0.6791\n",
            "Epoch 188/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0095 - accuracy: 0.6800\n",
            "Epoch 189/200\n",
            "20/20 [==============================] - 5s 227ms/step - loss: 0.0092 - accuracy: 0.6983\n",
            "Epoch 190/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0097 - accuracy: 0.6921\n",
            "Epoch 191/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0089 - accuracy: 0.7034\n",
            "Epoch 192/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0095 - accuracy: 0.7198\n",
            "Epoch 193/200\n",
            "20/20 [==============================] - 5s 232ms/step - loss: 0.0093 - accuracy: 0.6843\n",
            "Epoch 194/200\n",
            "20/20 [==============================] - 5s 238ms/step - loss: 0.0091 - accuracy: 0.6836\n",
            "Epoch 195/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0092 - accuracy: 0.6963\n",
            "Epoch 196/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0090 - accuracy: 0.6873\n",
            "Epoch 197/200\n",
            "20/20 [==============================] - 5s 227ms/step - loss: 0.0090 - accuracy: 0.7134\n",
            "Epoch 198/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0093 - accuracy: 0.6981\n",
            "Epoch 199/200\n",
            "20/20 [==============================] - 5s 234ms/step - loss: 0.0099 - accuracy: 0.6732\n",
            "Epoch 200/200\n",
            "20/20 [==============================] - 5s 233ms/step - loss: 0.0090 - accuracy: 0.7005\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "gc.collect()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:55:16.015932Z",
          "iopub.execute_input": "2022-08-01T20:55:16.016238Z",
          "iopub.status.idle": "2022-08-01T20:55:16.193073Z",
          "shell.execute_reply.started": "2022-08-01T20:55:16.016209Z",
          "shell.execute_reply": "2022-08-01T20:55:16.192033Z"
        },
        "trusted": true,
        "id": "kC47vdnF6OE7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "84078a35-d161-4671-c6e5-368a06ee87d3"
      },
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "50"
            ]
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder.load_weights('./model_auto.h5')"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:18.730569Z",
          "iopub.status.idle": "2022-08-01T20:39:18.731276Z",
          "shell.execute_reply.started": "2022-08-01T20:39:18.731022Z",
          "shell.execute_reply": "2022-08-01T20:39:18.731046Z"
        },
        "trusted": true,
        "id": "rU7WWBGW6OE7"
      },
      "execution_count": 62,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import keras.backend as K\n",
        "import tensorflow.keras.utils\n",
        "from tensorflow.keras.utils import plot_model\n",
        "from tensorflow.keras.layers import Layer, InputSpec\n",
        "\n",
        "class ClusteringLayer(Layer):\n",
        "    def __init__(self, n_clusters, weights=None, alpha=1.0, **kwargs):\n",
        "        if 'input_shape' not in kwargs and 'input_dim' in kwargs:\n",
        "            kwargs['input_shape'] = (kwargs.pop('input_dim'),)\n",
        "        super(ClusteringLayer, self).__init__(**kwargs)\n",
        "        self.n_clusters = n_clusters\n",
        "        self.alpha = alpha\n",
        "        self.initial_weights = weights\n",
        "        self.input_spec = InputSpec(ndim=2)\n",
        "\n",
        "    def build(self, input_shape):\n",
        "        assert len(input_shape) == 2\n",
        "        input_dim = input_shape[1]\n",
        "        self.input_spec = InputSpec(dtype=K.floatx(), shape=(None, input_dim))\n",
        "        self.clusters = self.add_weight(shape=(input_dim, self.n_clusters), initializer='glorot_uniform', name='clusters')\n",
        "        if self.initial_weights is not None:\n",
        "            self.set_weights(self.initial_weights)\n",
        "            del self.initial_weights\n",
        "        self.built = True\n",
        "\n",
        "    def call(self, inputs, **kwargs):\n",
        "        q = 1.0 / (1.0 + (K.sum(K.square(K.expand_dims(inputs, axis=1) - self.clusters), axis=2) / self.alpha))\n",
        "        q **= (self.alpha + 1.0) / 2.0\n",
        "        q = K.transpose(K.transpose(q) / K.sum(q, axis=1))\n",
        "        return q\n",
        "      \n",
        "    def compute_output_shape(self, input_shape):\n",
        "        assert input_shape and len(input_shape) == 2\n",
        "        return input_shape[0], self.n_clusters\n",
        "\n",
        "    def get_config(self):\n",
        "        config = {'n_clusters': self.n_clusters}\n",
        "        base_config = super(ClusteringLayer, self).get_config()\n",
        "        return dict(list(base_config.items()) + list(config.items()))"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:12:19.802298Z",
          "iopub.execute_input": "2022-08-01T21:12:19.802839Z",
          "iopub.status.idle": "2022-08-01T21:12:19.823319Z",
          "shell.execute_reply.started": "2022-08-01T21:12:19.802805Z",
          "shell.execute_reply": "2022-08-01T21:12:19.822339Z"
        },
        "trusted": true,
        "id": "Cu7JIb1W6OE8"
      },
      "execution_count": 63,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clustering_layer = ClusteringLayer(100, name='clustering')(encoder.output)\n",
        "clustering_model = Model(inputs=encoder.input, outputs=clustering_layer)\n",
        "clustering_model.compile(optimizer=SGD(0.01, 0.9), loss='kld')"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:12:23.172782Z",
          "iopub.execute_input": "2022-08-01T21:12:23.173438Z",
          "iopub.status.idle": "2022-08-01T21:12:23.284298Z",
          "shell.execute_reply.started": "2022-08-01T21:12:23.173401Z",
          "shell.execute_reply": "2022-08-01T21:12:23.283320Z"
        },
        "trusted": true,
        "id": "fHUVcdiH6OE8"
      },
      "execution_count": 64,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.cluster import KMeans\n",
        "kmeans = KMeans(n_clusters=100, n_init=20)\n",
        "y_pred = kmeans.fit_predict(encoder.predict(x))"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:12:28.837651Z",
          "iopub.execute_input": "2022-08-01T21:12:28.838751Z",
          "iopub.status.idle": "2022-08-01T21:12:45.861660Z",
          "shell.execute_reply.started": "2022-08-01T21:12:28.838714Z",
          "shell.execute_reply": "2022-08-01T21:12:45.860786Z"
        },
        "trusted": true,
        "id": "vwGprxS-6OE9"
      },
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clustering_model.get_layer(name='clustering').set_weights([kmeans.cluster_centers_])"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:12:46.222711Z",
          "iopub.execute_input": "2022-08-01T21:12:46.223390Z",
          "iopub.status.idle": "2022-08-01T21:12:46.233190Z",
          "shell.execute_reply.started": "2022-08-01T21:12:46.223353Z",
          "shell.execute_reply": "2022-08-01T21:12:46.232243Z"
        },
        "trusted": true,
        "id": "cjCxzygc6OE9"
      },
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "gc.collect()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:12:45.866384Z",
          "iopub.execute_input": "2022-08-01T21:12:45.869641Z",
          "iopub.status.idle": "2022-08-01T21:12:46.213454Z",
          "shell.execute_reply.started": "2022-08-01T21:12:45.869605Z",
          "shell.execute_reply": "2022-08-01T21:12:46.212503Z"
        },
        "trusted": true,
        "id": "9JGoL6jn6OE9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8a218b8d-dbb3-41f6-ecec-ca142ba45d1c"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "24519"
            ]
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred"
      ],
      "metadata": {
        "id": "3idnp4llYBSx",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ba4e8a22-1166-4c39-8e31-4759d3ed18a5"
      },
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 0,  0,  0, 79,  0,  0, 35, 32,  0, 83, 74, 29, 74,  4, 79, 45,  0,\n",
              "       79, 79,  0,  0, 57, 74, 74, 30,  0,  0,  0,  0,  0, 45,  0, 60, 59,\n",
              "        0,  0, 79,  0, 30,  0, 85,  0,  0, 79, 57,  0, 79, 70, 79, 74, 85,\n",
              "       79,  0, 74, 54,  0,  7, 79, 14, 85,  0, 41,  0, 60,  0, 30, 91, 92,\n",
              "       82, 30,  0, 69, 79, 35,  0,  4,  0,  0, 13, 74, 85, 15, 74,  0, 79,\n",
              "        0, 50,  5, 79, 33,  0, 12, 30,  8, 19, 79, 74, 79, 59, 46,  0,  8,\n",
              "        0, 33, 91,  0, 17, 79,  0, 89,  0, 79, 54, 30, 39,  0,  8, 29, 79,\n",
              "        0, 92, 77,  0,  4,  0,  0, 79, 13, 30,  0, 85,  0,  0, 74, 35, 79,\n",
              "        0, 79,  0, 74,  0,  0,  0,  7, 79, 35, 79,  0, 79,  0,  0,  0,  0,\n",
              "        5, 79, 56,  0, 60, 47, 23,  0, 79, 31, 79, 79,  0,  0,  0,  0, 35,\n",
              "       13,  9, 79, 29,  0, 79,  7, 30, 14,  0, 74, 54,  0, 73,  0,  0, 39,\n",
              "        0, 79, 74,  0, 36, 33, 60,  0, 79,  0, 13, 46,  0,  0,  0,  0, 79,\n",
              "       33, 79,  8,  0, 17,  0,  0,  0,  0,  0, 60, 31, 79,  0, 35, 74,  0,\n",
              "        0, 13, 79,  2,  0, 33, 73, 11,  0, 60, 79, 74,  0, 79,  0, 25,  0,\n",
              "       79, 67, 79, 39, 59,  0, 74,  0,  0,  8, 18, 79, 79, 79, 79,  0, 45,\n",
              "       58, 79, 49, 79,  0, 79,  0, 60,  0, 79,  0, 60,  0,  0, 79,  0, 74,\n",
              "        0,  0, 60, 33,  0,  0,  0, 25, 79,  0, 74, 74, 45,  0, 79, 79, 22,\n",
              "       94, 79,  0,  0, 13,  0,  0,  2,  0, 74,  0, 33,  0, 79,  0, 79, 35,\n",
              "        0,  0, 74,  0, 79, 79,  0,  0, 42, 58, 79,  0,  0, 12, 32, 17, 45,\n",
              "       79, 79,  0,  0,  0,  0, 16,  2, 45,  0,  0, 74, 26, 33,  0, 79,  0,\n",
              "       74, 58, 48, 74, 73, 88,  0,  0, 13, 85, 32, 79, 79,  0, 79, 13,  0,\n",
              "       82, 17,  7, 20,  8, 79, 60, 74,  0,  0,  0,  0, 74, 79,  0, 79,  0,\n",
              "        0, 33, 26, 17,  0, 79, 72,  0, 52,  4,  6, 49,  0, 60, 55,  3, 79,\n",
              "        0, 35, 45,  0,  0, 63,  0,  0, 79, 10, 79, 32, 33,  0, 79, 24, 32,\n",
              "        0, 79, 79, 79,  0, 37, 79,  2,  0,  0,  0, 79,  0, 13, 74, 82, 79,\n",
              "       79, 32, 35,  0, 13, 42, 73, 99, 13, 60, 79, 79, 79, 33, 79, 79,  0,\n",
              "        0, 79, 60, 35,  0,  0,  0, 79, 79, 79, 42,  0, 75,  0,  0,  0,  0,\n",
              "       50, 79,  0, 35, 60,  0, 83, 79, 13, 79, 30, 60,  0,  0, 74, 79, 33,\n",
              "        0, 79,  8,  0, 79, 79, 79,  0, 79, 79, 79, 74, 59, 59, 33, 30, 54,\n",
              "       79, 60, 79,  0,  0, 64, 61,  0, 54, 54,  0,  0, 32, 30,  0,  0,  0,\n",
              "       60, 79, 21,  5, 79,  0,  0, 17,  0,  0,  5, 91, 79, 54, 18,  5, 74,\n",
              "       30,  0, 28,  0, 79,  0,  0,  0, 33, 79,  0,  0, 29,  0, 20, 74,  0,\n",
              "        0,  1, 74, 79,  0,  5, 30, 76,  0, 60,  0,  0,  0,  0, 74,  0, 79,\n",
              "        0,  0,  0, 13, 33, 15, 35,  0,  0, 85, 79, 74,  0,  0,  0,  0, 63,\n",
              "       13,  9,  0,  0, 12, 60, 60, 79, 74, 79, 79, 79,  0, 79, 54,  0, 85,\n",
              "       30, 74, 33,  0, 79, 60, 79,  0, 62, 33,  0, 81,  0, 79,  7, 60,  0,\n",
              "        0, 42, 46,  0, 58,  0, 79,  0,  0, 48, 95, 79, 58,  0, 18, 35,  0,\n",
              "       33, 79, 74, 87, 38, 33,  0,  0,  0, 79,  0,  0, 79, 21, 13,  0, 13,\n",
              "       58, 79, 86,  0,  0,  8,  0, 91,  0, 74,  8,  0,  0,  0, 74, 74,  0,\n",
              "       79, 59,  0, 60, 79, 74, 39, 58, 62, 42, 79, 96,  0,  0, 48,  0,  0,\n",
              "       79,  0,  0, 60, 91, 79, 74,  0, 64, 60,  7,  0, 11, 58, 26,  0,  0,\n",
              "       19,  0,  0, 30, 79,  4, 79, 64, 79, 33,  0, 30, 54, 79,  0,  0,  0,\n",
              "        0, 46,  0, 93, 79, 79, 79, 60, 29, 89,  8,  0,  0,  0, 74, 60,  0,\n",
              "        0, 79, 74, 73,  0, 13, 79,  0,  0, 52, 39, 80,  0,  0, 79, 65,  0,\n",
              "       97, 79, 74,  0, 79, 35,  0, 14, 79, 27, 79, 20,  0, 33, 42,  0, 35,\n",
              "       33,  0,  0, 79, 60,  0,  0,  0, 15, 79,  0,  0, 39,  0, 28, 11,  0,\n",
              "       67,  0, 33,  0, 44, 79,  0, 79, 79,  0, 91, 79, 74,  0,  0, 79,  0,\n",
              "        0,  0, 81, 68, 84,  0,  0,  0,  0, 60,  0,  0,  0,  0,  8, 26, 45,\n",
              "       43,  0, 29,  0, 79, 79,  0, 60, 79,  0,  8, 60, 79,  0, 92, 79, 13,\n",
              "        0, 79, 79, 32, 60, 72, 60,  0,  0,  0, 79, 79,  0,  0,  0, 57, 51,\n",
              "       79, 33, 79,  0,  0,  0, 54, 31,  0,  0, 73,  0,  0, 58, 79, 13,  0,\n",
              "       74, 49,  0, 13, 60, 60, 79, 79,  0, 40,  0,  0,  0,  0,  0,  0,  0,\n",
              "        0, 74, 90, 79,  0, 79, 79,  0, 33, 45, 34,  0, 79, 79,  0, 13,  0,\n",
              "       35, 74, 59, 52, 89, 79, 35, 46,  0, 79,  0,  0,  0, 74, 30, 79, 30,\n",
              "       25,  0,  0, 58, 32,  0, 74, 78,  0,  0, 60, 79,  0, 64,  0, 53,  0,\n",
              "        0, 66, 79, 33, 60,  2, 74, 79, 79, 79, 79, 13, 74, 60,  0,  0, 30,\n",
              "       91, 79, 67,  0, 67,  8, 58,  4, 79, 36,  8, 98,  0, 30,  0,  0, 34,\n",
              "        8, 74, 67, 11, 74, 79, 74, 79, 54,  0, 33, 79, 79, 50,  0, 71,  0,\n",
              "        0,  8,  0, 37, 17,  0,  0,  0,  8,  0, 79, 60,  0], dtype=int32)"
            ]
          },
          "metadata": {},
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\"q = clustering_model.predict(x, verbose=0)\""
      ],
      "metadata": {
        "id": "pGjx4ToyY--T",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "outputId": "72c0158d-61ab-45ba-a44d-806a8580933d"
      },
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'q = clustering_model.predict(x, verbose=0)'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\"p = target_distribution(q)  # update the auxiliary target distribution p\""
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:20:04.403291Z",
          "iopub.execute_input": "2022-08-01T21:20:04.404786Z",
          "iopub.status.idle": "2022-08-01T21:20:20.360836Z",
          "shell.execute_reply.started": "2022-08-01T21:20:04.404743Z",
          "shell.execute_reply": "2022-08-01T21:20:20.359554Z"
        },
        "trusted": true,
        "id": "LojQ_WOw6OFD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "outputId": "d3c313f3-f094-4003-92f5-3553c4baa10a"
      },
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'p = target_distribution(q)  # update the auxiliary target distribution p'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\"y_pred = q.argmax(1)\""
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:20:20.363013Z",
          "iopub.execute_input": "2022-08-01T21:20:20.363400Z",
          "iopub.status.idle": "2022-08-01T21:20:20.368347Z",
          "shell.execute_reply.started": "2022-08-01T21:20:20.363362Z",
          "shell.execute_reply": "2022-08-01T21:20:20.367421Z"
        },
        "trusted": true,
        "id": "Hp1cXFz06OFD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "outputId": "8c3969e9-06a2-44d8-bb1d-773a8e58872d"
      },
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'y_pred = q.argmax(1)'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\"y_pred\""
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:20:20.370983Z",
          "iopub.execute_input": "2022-08-01T21:20:20.371740Z",
          "iopub.status.idle": "2022-08-01T21:20:20.383464Z",
          "shell.execute_reply.started": "2022-08-01T21:20:20.371702Z",
          "shell.execute_reply": "2022-08-01T21:20:20.382456Z"
        },
        "trusted": true,
        "id": "nYhOq4hr6OFD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "outputId": "777faa5c-db32-455c-b0fc-14c3399f42a7"
      },
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'y_pred'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "gc.collect()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:39:23.356211Z",
          "iopub.execute_input": "2022-08-01T21:39:23.357427Z",
          "iopub.status.idle": "2022-08-01T21:39:23.709184Z",
          "shell.execute_reply.started": "2022-08-01T21:39:23.357381Z",
          "shell.execute_reply": "2022-08-01T21:39:23.708199Z"
        },
        "trusted": true,
        "id": "Zio8DxrQ6OFD",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "31baf7cf-6dcd-4a10-db01-de53172aa95e"
      },
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "300"
            ]
          },
          "metadata": {},
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\"\"\"for i in range(100):\n",
        "    idx = index_array[index * batch_size: min((index+1) * batch_size, x.shape[0])]\n",
        "    loss =clustering_model.train_on_batch(x=x[idx], y=p[idx])\n",
        "    index = index + 1 if (index + 1) * batch_size <= x.shape[0] else 0\n",
        "    gc.collect()\n",
        "print(loss)\"\"\""
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:19:13.919445Z",
          "iopub.execute_input": "2022-08-01T21:19:13.920156Z",
          "iopub.status.idle": "2022-08-01T21:20:04.401358Z",
          "shell.execute_reply.started": "2022-08-01T21:19:13.920119Z",
          "shell.execute_reply": "2022-08-01T21:20:04.400268Z"
        },
        "trusted": true,
        "id": "pta0E3WU6OFD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "outputId": "7b643254-bd5a-470d-8f1e-857a9cb465af"
      },
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'for i in range(100):\\n    idx = index_array[index * batch_size: min((index+1) * batch_size, x.shape[0])]\\n    loss =clustering_model.train_on_batch(x=x[idx], y=p[idx])\\n    index = index + 1 if (index + 1) * batch_size <= x.shape[0] else 0\\n    gc.collect()\\nprint(loss)'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 74
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def target_distribution(q):\n",
        "    weight = q ** 2 / q.sum(0)\n",
        "    return (weight.T / weight.sum(1)).T"
      ],
      "metadata": {
        "id": "xYEuK0PXT4L6",
        "execution": {
          "iopub.status.busy": "2022-08-01T21:39:01.712869Z",
          "iopub.execute_input": "2022-08-01T21:39:01.713247Z",
          "iopub.status.idle": "2022-08-01T21:39:01.720892Z",
          "shell.execute_reply.started": "2022-08-01T21:39:01.713215Z",
          "shell.execute_reply": "2022-08-01T21:39:01.717786Z"
        },
        "trusted": true
      },
      "execution_count": 75,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "loss = 0\n",
        "index = 0\n",
        "maxiter = 10000\n",
        "update_interval = 100\n",
        "index_array = np.arange(x.shape[0])\n",
        "batch_size = 16"
      ],
      "metadata": {
        "id": "lwrpzzM5YYQb"
      },
      "execution_count": 76,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for ite in range(int(maxiter)):\n",
        "    if ite % update_interval == 0:\n",
        "        q = clustering_model.predict(x, verbose=0)\n",
        "        p = target_distribution(q)  # update the auxiliary target distribution p\n",
        "        gc.collect()\n",
        "\n",
        "        from sklearn.manifold import TSNE\n",
        "        x_embedded = TSNE(n_components=2).fit_transform(p)\n",
        "        vis_x = x_embedded[:, 0]\n",
        "        vis_y = x_embedded[:, 1]\n",
        "        plt.scatter(vis_x, vis_y, c=y_pred, cmap=plt.cm.get_cmap(\"jet\", 256))\n",
        "        plt.colorbar(ticks=range(256))\n",
        "        plt.clim(-0.5, 9.5)\n",
        "        plt.show()\n",
        "        # evaluate the clustering performance\n",
        "        y_pred = q.argmax(1)\n",
        "        print(\"prediction \",y_pred)\n",
        "        print(\"iter \",ite,\" loss \",loss)\n",
        "    idx = index_array[index * batch_size: min((index+1) * batch_size, x.shape[0])]\n",
        "    loss =clustering_model.train_on_batch(x=x[idx], y=p[idx])\n",
        "    index = index + 1 if (index + 1) * batch_size <= x.shape[0] else 0\n",
        "    gc.collect()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T21:40:08.231154Z",
          "iopub.execute_input": "2022-08-01T21:40:08.231534Z",
          "iopub.status.idle": "2022-08-01T21:54:41.207372Z",
          "shell.execute_reply.started": "2022-08-01T21:40:08.231484Z",
          "shell.execute_reply": "2022-08-01T21:54:41.205999Z"
        },
        "trusted": true,
        "id": "VR8kElrO6OFE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "b2d3356f-b586-4f06-ad99-edee446f2c4f"
      },
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [ 0  0  0 79  0  0 35 32  0 83 74 29 74  4 79 45  0 79 79  0  0 57 74 74\n",
            " 30  0  0  0  0  0 45  0 60 59  0  0 79  0 30  0 85  0  0 79 57  0 79 70\n",
            " 79 74 85 79  0 74 54  0  7 79 14 85  0 41  0 60  0 30 91 92 82 30  0 69\n",
            " 79 35  0  4  0  0 13 74 85 15 74  0 79  0 50  5 79 33  0 12 30  8 19 79\n",
            " 74 79 59 46  0  8  0 33 91  0 17 79  0 89  0 79 54 30 39  0  8 29 79  0\n",
            " 92 77  0  4  0  0 79 13 30  0 85  0  0 74 35 79  0 79  0 74  0  0  0  7\n",
            " 79 35 79  0 79  0  0  0  0  5 79 56  0 60 47 23  0 79 31 79 79  0  0  0\n",
            "  0 35 13  9 79 29  0 79  7 30 14  0 74 54  0 73  0  0 39  0 79 74  0 36\n",
            " 33 60  0 79  0 13 46  0  0  0  0 79 33 79  8  0 17  0  0  0  0  0 60 31\n",
            " 79  0 35 74  0  0 13 79  2  0 33 73 11  0 60 79 74  0 79  0 25  0 79 67\n",
            " 79 39 59  0 74  0  0  8 18 79 79 79 79  0 45 58 79 49 79  0 79  0 60  0\n",
            " 79  0 60  0  0 79  0 74  0  0 60 33  0  0  0 25 79  0 74 74 45  0 79 79\n",
            " 22 94 79  0  0 13  0  0  2  0 74  0 33  0 79  0 79 35  0  0 74  0 79 79\n",
            "  0  0 42 58 79  0  0 12 32 17 45 79 79  0  0  0  0 16  2 45  0  0 74 26\n",
            " 33  0 79  0 74 58 48 74 73 88  0  0 13 85 32 79 79  0 79 13  0 82 17  7\n",
            " 20  8 79 60 74  0  0  0  0 74 79  0 79  0  0 33 26 17  0 79 72  0 52  4\n",
            "  6 49  0 60 55  3 79  0 35 45  0  0 63  0  0 79 10 79 32 33  0 79 24 32\n",
            "  0 79 79 79  0 37 79  2  0  0  0 79  0 13 74 82 79 79 32 35  0 13 42 73\n",
            " 99 13 60 79 79 79 33 79 79  0  0 79 60 35  0  0  0 79 79 79 42  0 75  0\n",
            "  0  0  0 50 79  0 35 60  0 83 79 13 79 30 60  0  0 74 79 33  0 79  8  0\n",
            " 79 79 79  0 79 79 79 74 59 59 33 30 54 79 60 79  0  0 64 61  0 54 54  0\n",
            "  0 32 30  0  0  0 60 79 21  5 79  0  0 17  0  0  5 91 79 54 18  5 74 30\n",
            "  0 28  0 79  0  0  0 33 79  0  0 29  0 20 74  0  0  1 74 79  0  5 30 76\n",
            "  0 60  0  0  0  0 74  0 79  0  0  0 13 33 15 35  0  0 85 79 74  0  0  0\n",
            "  0 63 13  9  0  0 12 60 60 79 74 79 79 79  0 79 54  0 85 30 74 33  0 79\n",
            " 60 79  0 62 33  0 81  0 79  7 60  0  0 42 46  0 58  0 79  0  0 48 95 79\n",
            " 58  0 18 35  0 33 79 74 87 38 33  0  0  0 79  0  0 79 21 13  0 13 58 79\n",
            " 86  0  0  8  0 91  0 74  8  0  0  0 74 74  0 79 59  0 60 79 74 39 58 62\n",
            " 42 79 96  0  0 48  0  0 79  0  0 60 91 79 74  0 64 60  7  0 11 58 26  0\n",
            "  0 19  0  0 30 79  4 79 64 79 33  0 30 54 79  0  0  0  0 46  0 93 79 79\n",
            " 79 60 29 89  8  0  0  0 74 60  0  0 79 74 73  0 13 79  0  0 52 39 80  0\n",
            "  0 79 65  0 97 79 74  0 79 35  0 14 79 27 79 20  0 33 42  0 35 33  0  0\n",
            " 79 60  0  0  0 15 79  0  0 39  0 28 11  0 67  0 33  0 44 79  0 79 79  0\n",
            " 91 79 74  0  0 79  0  0  0 81 68 84  0  0  0  0 60  0  0  0  0  8 26 45\n",
            " 43  0 29  0 79 79  0 60 79  0  8 60 79  0 92 79 13  0 79 79 32 60 72 60\n",
            "  0  0  0 79 79  0  0  0 57 51 79 33 79  0  0  0 54 31  0  0 73  0  0 58\n",
            " 79 13  0 74 49  0 13 60 60 79 79  0 40  0  0  0  0  0  0  0  0 74 90 79\n",
            "  0 79 79  0 33 45 34  0 79 79  0 13  0 35 74 59 52 89 79 35 46  0 79  0\n",
            "  0  0 74 30 79 30 25  0  0 58 32  0 74 78  0  0 60 79  0 64  0 53  0  0\n",
            " 66 79 33 60  2 74 79 79 79 79 13 74 60  0  0 30 91 79 67  0 67  8 58  4\n",
            " 79 36  8 98  0 30  0  0 34  8 74 67 11 74 79 74 79 54  0 33 79 79 50  0\n",
            " 71  0  0  8  0 37 17  0  0  0  8  0 79 60  0]\n",
            "iter  0  loss  0\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [ 0  0  0  0  0  0  0  0  0  0  0  0  0 30  0  0  0 79 33  0  0  0  0  0\n",
            "  4  0  0  0  0  0  0  0  0  0  0  0  0  0 13  0  0  0  0 79 60  0 79  0\n",
            "  0  0  0  0  0  0 54  0  0  0  0  0  0  0  0 79  0 79  0 18 82 30  0  0\n",
            " 79  0  0 30  0  0 13  0  0  0  0  0  0  0 60  0  0 33  0  0 60  8  0  0\n",
            "  0  0  0 74  0  4  0 79  0  0 79  0  0  0  0 79 54 79  0 79  8  0 79  0\n",
            " 92 74  0  4  0  0 79 30  0  0  0  0  0  0  0 79  0 79  0  0  0  0  0 60\n",
            " 79  0 79  0  0  0  0  0  0  0  0  0  0 79 74  0  0  0  0  0  0  0 79  0\n",
            "  0  0 30  0  0  0  0 79  0 33  0  0  0 33  0  0  0  0  0  0  0  0  0  0\n",
            " 79 79  0  0  0 33  0  0  0  0  0  0 79  0 54  0  0  0  0  0  0  0  0  0\n",
            " 79  0  0  0  0  0 60  0 74  0 79  0  0  0  0  0  0  0  0  0  0  0 79  0\n",
            " 79  0  0  0  0  0  0 54 18 79  0 79  0  0  0  0  0  0  0  0 79  0 79  0\n",
            "  0  0  0  0  0  0  0  0  0  0  0 79  0  0  0  0 79  0  0  0  0  0 79  0\n",
            "  0  0  0  0  0 33  0  0  0  0  0  0 33  0  0  0 79  0  0  0  0  0 79  0\n",
            "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 79  0 48\n",
            " 79  0  0  0  0  0 48  0  0  0  0  0 60  0  0  0  0  0  0 79  0 82  0  0\n",
            "  0 54  0  0  0  0  0  0  0  0  0  0  0  0  0 79 48  0  0  0  0  0  0 30\n",
            " 74 74  0  0  0  0 79  0  0  0  0 79  0  0  0  0  0  0 74 79  0 79 74  0\n",
            "  0  0 79  0  0  0  0  0  0  0  0  0  0 79  0 82  0  0  0  0  0 33  0  0\n",
            "  0 33  0 79  0  0 79  0  0  0  0 79 60  0  0  0  0  0  0  0  0  0  0  0\n",
            "  0  0  0 60 79  0  0  0  0  0  0 30  0 79 79  0  0  0  0 33  0  0 30  0\n",
            "  0  0  0  0  0  0  0  0  0  0  0 60 54  0  0 79  0  0  0  0  0 54 54  0\n",
            "  0 74 79  0  0  0 79  0  0  0  0  0  0  0  0  0  0  0  0 30 18  0  0 79\n",
            "  0  0  0 79  0  0  0 33  0  0  0  0  0  0  0  0  0  0  0  0  0  0 79  0\n",
            "  0 79  0  0  0  0  0  0  0  0  0  0 60 79  0 79  0  0  0  0  0 79  0  0\n",
            "  0  0 33  0  0  0  0  0 79  0  0 79  0 79  0  0 54  0  0 33  0  0  0  0\n",
            " 60  0  0  0 79  0  0  0  0  0 79  0  0  0  0  0  0  0 79  0  0 48  0 79\n",
            " 74  0 82  0  0  0 79  0  0  0 33  0  0  0 79  0  0  0 74 13  0 30  0  0\n",
            "  0  0  0 30  0  0  0  0  8  0  0  0  0  0  0  0  0 60 60  0  0  0  0  0\n",
            "  0 79  0 30  0 48  0  0  0  0  0  0  0  0  0  0  0 79  0  0  0  0 26 79\n",
            "  0  0  0  0 79  0  4  0  0 79 33  0 13 33  0  0  0  0  0  0  0  0  0 79\n",
            "  0 79  0  0 18  0  0  0  0 60  0  0  0  0  0  0 30  0  0  0  0  0  0  0\n",
            "  0  0  0  0  0 79  0  0  0  0  0  0 79  0 79 74 79 79  0  0  0 33  0  0\n",
            "  0 60  0  0  0  0  0  0  0  0  0  0  0  0  0  0 33 79 74  0  0  0  0  0\n",
            "  0 79  0  0  0  0 79  0 79 79  0  0  0  0  0  0  0  0  0  0  0 13 26  0\n",
            "  0  0  0  0  0 79  0 79 79  0 30 79  0  0 92  0 79 79  0  0  0  0  0 79\n",
            "  0  0  0  0  0  0  0  0  0  0  0 79  0  0  0  0 54  0 79  0  0  0  0  0\n",
            "  0 79  0  0  0  0 54 60 60  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "  0  0  0  0 79  0  0  0  0  0  0 30  0  0  0  0  0  0 79  0  0  0  0  0\n",
            "  0  0  0 33 79 79  0  0  0  0  0  0  0 74 79  0 79 79  0  0  0  0  0  0\n",
            "  0  0 79 79  0  0  0 79  0  0 30  0 79  0  0 79  0  0  0  0  0 13  0 30\n",
            " 79  0 26  0  0 33  0  0  0 13 79  0  0  0  0  0 79 33  0 33  0  0 60  0\n",
            "  0  0  0  8  0  0  0  0  0  0  8  0 79 60  0]\n",
            "iter  100  loss  1.1468387842178345\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [ 0  0 79 79  0  0  0  0  0  0 79  0  0 26 79  0 79 33 92  0 79  0  0  0\n",
            " 26  0  0  0  0  0  0 79 33  0  0  0 79  0 18  0  0 79  0 33 79 79 79 74\n",
            " 79  0  0 79  0 79 92  0  0 79 74  0  0 74  0 33  0 13  0 18 82  8  0  0\n",
            " 33 60  0 26 79 79 18  0  0  0 79  0 79 79 60  0 79 54 79 74 13 82 74 79\n",
            "  0 79  0  0 79 48  0 33  0 79 33 79 79  0  0 33 92 13  0 33 82  0 54 79\n",
            " 18 74 79 48  0  0 33 26 13  0  0 79  0 79  0 33 79 33 79  0  0 79 79 30\n",
            " 79  0 54  0 79 79 79 79  0  0 79  0  0 30 74  0  0 33  0  0 79  0 79  0\n",
            " 60  0  8  0 79  0 79 33  0  8 74  0  0 54 79  0 79 79  0  0 79  0 79 60\n",
            " 33 60  0 79 79  8 74 79 79 79 79 79 33 33 26 79 74 79 79  0  0  0 33 74\n",
            " 79  0  0  0  0  0  8 79 74  0 54  0 74 79 33 79  0  0 79 79  0 79 33  0\n",
            " 33 74  0  0  0 79 79 18 18 33 79 33 79  0  0 74 79 74 33 79 33  0 33  0\n",
            " 79  0 33  0  0 33  0  0  0  0 13 33 79 79 79 30 33  0  0  0  0 79 79 79\n",
            "  0 74 79  0  0  8  0 79  0  0  0 79 54  0 79  0 33  0  0 79  0  0 33 79\n",
            "  0 79  0  0 79  0  0  0  0  0  0 79 60  0  0  0 79 30  0  0  0 33  0 48\n",
            " 54 79 33  0  0  0 48  0 79  0 79  0  8  0  0 79 33  0 79  8 79 82  0  0\n",
            "  0 18 79 79  0 79  0  0  0 74 33 79 79 60  0 54 48  0  0 79  0 79  0 26\n",
            " 74 74 79 33 74  0 33 79  0  0  0 79  0  0 79 33  0 79  0 54  0 33  0  0\n",
            " 79 79 33 79  0 74 79  0 79  0 79 79 79  8  0 82 79 79 74  0  0  8  0  0\n",
            "  0  8 33 33 79 79 33 79 79  0 79 79 30  0 79  0 79 79 79 79 74  0 74 79\n",
            " 79 79 79  4 33 79  0 30 79  0 79 18 79 54 33  0  0  0 79 54 79 79 82 79\n",
            " 79 79 79 79 79 79 79  0 74  0 33  8 92 79 30 33  0 79  0  0 74 92 18  0\n",
            "  0 74 33 79  0  0 13 79  0  0 79 79 79  0  0 79  0  0 79 54 18  0  0 13\n",
            "  0  0 79 33  0  0 79 54 79  0 79  0  0 74  0 79  0 74  0 79  0 74 33  0\n",
            "  0 33  0  0  0  0  0  0 79 60  0 79  8 54 74 33  0  0  0 79  0 33  0  0\n",
            "  0 74  8  0 79  0 74 33 33 33  0 33 33 33  0 79 92  0  0 13 79 54  0 79\n",
            " 13 79  0 74 33 79 79  0 79  0 33  0  0 74  0  0  0  0 79 79 79 48  0 33\n",
            "  0  0 18  0 79 33 33  0 74 74 54  0  0 79 33  0  0 79 74 26  0 54  0 79\n",
            "  0  0  0 82 79 74  0  0 48  0  0 79  0  0  0 79  0 54 13 79 79  0  0 74\n",
            " 74 30 74 13 79 48  0 33 79  0 79 13  0 79 60  0 74 33  0 79  0  0 48 79\n",
            " 79 74 79  0 13 33 26 79 79 79 54 79 18 92 79  0  0  0  0  0 79  0 79 33\n",
            " 79 13  0  0 82 79  0 79  0 13  0  0 33  0  0 79  8 79 79 79  0  0  0  0\n",
            "  0 79 74  0  0 33  0  0 79  0  0 74 79  0 33  0 33 54 74  0  0 54  0  0\n",
            " 79 54 79  0  0  0 79  0 79 74  0 74 74  0 74  0 54 33  0 79 79 79 79 79\n",
            "  0 33  0 79  0 79 33 79 33 30  0  0  0 79 79 79 60 79  0 79 79 82 48  0\n",
            "  0  0  0 79 79 79 79 30 33 79 26 33 79 79 18 33  8 33 33 79  0 30 74 33\n",
            " 33  0  0 79 79 79  0 79 60  0 79 54 79  0  0 79 92  0 30 60  0  0  0  0\n",
            " 79  8 79  0 74 79 18 30 13 79 79  0  0  0  0  0  0 79  0 79  0  4 74 79\n",
            " 79 33 79 79 33  0 74 79 79 79 60  8 79  0  0  0  0  0 33  0 74 79 79  0\n",
            "  0  0  0 33 33 30 30  0  0 74  0 79  0 74 33  0 33 33 79 74  0  0 79  0\n",
            " 74 33 54 33 74  0 33 33 79 79  8  0 13  0  0 54 74 79 74  0  0 82 74 26\n",
            " 33 74 48 74  0 54 79 79  0 82 33 74  0 79 79  0 33 54  0 54 30 79  8 79\n",
            "  0 79  0 48 79 74 74  0  0 79 48  0 33 30  0]\n",
            "iter  200  loss  0.3190864324569702\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [ 0 79  0  0  0  0  0  0  0  0  0  0 74  8 79 74  0 33 92  0 79  0  0  0\n",
            "  8  0  0  0  0  0  0  0 33  0  0  0  0  0  8  0 74  0  0 79 60 79 79 74\n",
            " 79  0 74 79  0 79 18  0  0 79 74 74  0 74  0 33  0 13  0 54 82  8  0  0\n",
            " 33  0  0 26  0  0 18 74  0  0 79  0 79  0 30  0 79 54 79 74 13 48  0  0\n",
            "  0  0  0 74 79 48  0 33  0  0 33  0 79  0  0 33 92 13  0 33 82  0 54  0\n",
            " 18 74 79 48  0  0 33 26 30  0  0 79  0  0  0 33 79 33 79  0  0 79 79 30\n",
            " 79 74 54  0  0  0 79  0  0 74 60 74  0 33 74  0  0 79 74  0 79  0 79  0\n",
            "  0  0 26  0  0  0 79 33  0  8 74  0 74 54 79  0 79 79 74  0  0 74  0 74\n",
            " 33 60  0 79 79  8  0  0 79  0  0 33 33 33 82  0 74  0 79  0  0  0 79 74\n",
            " 79  0  0  0 74  0  8 79 74  0 54  0 74  0 79  0 74  0 79  0  0  0 33  0\n",
            " 33 74 74  0  0  0  0 82 82 33 79 33  0  0  0 74  0 74 33  0 33  0 33  0\n",
            " 79  0 79  0  0 79  0  0  0  0 13 54 79 79  0 60 33  0  0  0 74  0 79 79\n",
            " 60 74 79  0  0  8  0  0 74  0  0  0 54  0 79 74 79 74  0  0  0  0 54  0\n",
            "  0  0 74  0 79  0  0  0 74  0 74 79 60  0  0  0  0 30 74  0  0 79  0 48\n",
            " 54 79 33  0  0  0 48 74  0 74  0  0  8  0 74 79 33  0 79  8 79 82  0  0\n",
            "  0 18 79 79  0 79  0  0  0 74 79  0 79 79  0 54 48  0  0 79 74  0 74 26\n",
            " 74 74  0 79 74  0 33  0 74 74  0 79 74  0  0 79 74  0 74 33  0 33 74 74\n",
            "  0 79 33  0  0  0 79 74  0  0 79 79  0  8  0 54 79 79 74  0  0  8  0  0\n",
            " 74  8 60 79 79 79 33  0 79  0 79 79 30 74  0  0 79 79 79 79 74  0  0 79\n",
            "  0 79 79  4 79 79  0 60 79  0 79 26 79 13 60  0  0  0 79 54  0 79 26  0\n",
            " 79 79 79  0 79 79 33  0 74  0 79  8 79 79 30 33  0  0  0 74 74 18 18  0\n",
            "  0 74 60  0  0  0 13 79 74  0  0 74  0  0  0  0 74 74  0 54 18 74 74 13\n",
            "  0  0 79 33 74  0  0 54 79  0 79  0  0  0  0 79  0 74 74  0  0 74 33 74\n",
            "  0 33  0 74  0  0 79  0 79  0  0  0 26 54 74 33  0 74 74  0  0 33  0 74\n",
            "  0 74  8  0 79  0  0 79 33 79  0 33 79 33  0 79 92  0 79  8 79 54  0  0\n",
            " 13  0  0  0 33  0 79  0 79  0 33 74  0 74 74  0  0  0 79  0  0 48  0 33\n",
            " 74  0 82  0  0 54 33 74 74 74 54  0  0 74 33  0  0 79 74 26  0 54  0 79\n",
            "  0  0  0 26  0 74  0 74 48  0 74  0  0  0  0 79  0 54 13  0 74  0 74 74\n",
            " 74 30 74 13 79 82  0 33 79  0 79 30 74 79 79  0 74 33  0  0 74  0 48 74\n",
            " 74  0 79  0  8 33 48  0 33 79 92  0 82 18 79  0  0  0  0  0  0 74 79 33\n",
            " 79 13  0  0 82 79  0 79  0 13  0  0 33  0 79 79  8 79 79  0 74 74 74  0\n",
            "  0 79  0  0  0 33  0  0  0  0  0 74 79  0 33 74 79 54 74  0 74 54  0  0\n",
            "  0 54  0  0  0 74  0  0  0  0  0  0 74  0 74  0 54 54  0 79 79 79 79 79\n",
            " 74 33  0 79  0 79 33  0 33 30 79  0  0 79  0 79 60  0  0  0  0 48 48 74\n",
            "  0  0  0  0 79 79 79 30 33  0 26 79  0 79 18 33  8 33 33 79 74 30 74 79\n",
            " 79  0  0 79  0  0  0  0 30 74 79 54 79  0  0 79 92 74 60 79  0  0  0  0\n",
            " 79  8 79  0 74  0 18 30 13  0 79  0  0  0  0  0  0  0  0 79  0 30 74  0\n",
            "  0 79  0  0 79 74 74  0 79  0 79  8  0  0  0 74 74  0 33 74 74  0  0 74\n",
            "  0  0  0 33 33 33 30  0 74  0  0 79 74 74 33  0 33 79 79 74 74  0 79  0\n",
            "  0 33 33 33 74  0 79 33 79 60  8  0 13  0 79 54 74 79 74  0  0 48  0 26\n",
            " 79  0 48 74  0 92  0  0 74 82 33 74 74 79 79  0 33 92  0 92 30 79  8  0\n",
            "  0  0  0 48  0 74 74  0  0  0 48  0 33 30  0]\n",
            "iter  300  loss  0.16230155527591705\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [ 0 79 79  0  0  0  0  0  0  0  0  0 74 60 79 74  0 33 54  0  0  0  0  0\n",
            " 30  0  0  0  0  0  0 79 79  0  0  0 79  0 33  0 74 79  0  0  0 79 79 74\n",
            " 79  0 74 79  0  0 54  0  0 33  0 74  0 35  0 33  0 79  0  0 92 30  0 74\n",
            " 33  0  0 30  0 79 54 74  0  0  0  0 79 79  0 74 33 79 79  0 79  8  0 79\n",
            "  0 79 74 74 33  8  0 33  0  0 79  0  0  0  0 33 54 79  0 79  8  0 33 79\n",
            " 92 74 79  8  0  0 33 30 79  0  0 79  0  0  0 33 79 33 79  0  0 33 79  0\n",
            " 79 74 33  0 79  0 79 79  0 74  0  0  0 33 74  0  0 79 74 79 79  0 79  0\n",
            "  0  0 60  0  0  0  0 33 74 79 74  0 74 33 79  0 79 79  0  0 79  0 79 74\n",
            " 33 33  0 33 79 33  0  0 79  0  0 33 33 79 30  0 74  0 79  0  0  0 79 74\n",
            " 79  0 74  0 74  0 79 79 74  0 54  0 74  0 79  0 74  0 79 79  0 79 79  0\n",
            " 33 74 74  0  0 79  0 33 92 33 79 79  0  0  0 74  0 74 33  0 33 74 79  0\n",
            "  0  0 79 79  0 33  0  0  0 79 79 54 79 33 79  0 33 74  0  0 74 79 79 79\n",
            "  0 35 33  0  0 54  0  0 74  0  0  0 33  0 33 74 33 74  0 79  0  0 33 79\n",
            "  0 79  0  0 79  0  0  0 74  0 74 79  0  0  0  0  0  0 74  0  0 33  0 48\n",
            " 33 79 33  0  0  0 48 74  0  0  0  0 33  0 74 33 33  0 79 33 79 18 79 74\n",
            "  0 54 79 33  0 79  0  0  0 74 33  0 33 79  0 33 48  0  0 33  0  0 74 30\n",
            " 35 74  0 79 74 74 79  0 74 74  0 79  0  0  0 33  0 79 74 33  0 33 74 74\n",
            " 79 33 33  0  0  0 33 74 79  0 79 79  0 33  0 74 79 79  0  0  0 33 74 74\n",
            " 74 33 79 33 79 79 79 79 79  0 79 79 79 74 79  0 79 33 79 33 74  0 74 79\n",
            " 79 79  0  0 33  0  0 79 79  0  0 33 79 79 33  0  0  0 79 33 79 79 54 79\n",
            " 79 79 79 79 79 79 33  0 35 74 79 33  0 79  0 33  0  0  0 74 74 54 92  0\n",
            "  0 74 79 79  0  0 79 79  0 74 79 74 79 74  0 79 74 74  0 33 33 74 74 79\n",
            "  0 74 79 54 74  0  0 54 79  0 79  0  0  0  0  0  0 74 74  0  0 74 33  0\n",
            "  0 33  0 74  0  0 79  0 79  0  0 79 30 33  0 33  0 74 74  0  0 33  0 74\n",
            "  0 74 30 74 79  0 74 79 33 33  0 33 33 33  0 79 54  0 79 79 79 79  0  0\n",
            " 33  0  0 74 33 79  0  0 79 74 33 74  0 74 74  0 74  0 79 79  0 48  0 33\n",
            " 74  0 92  0 79 79 33  0  0 74 33  0  0  0 33  0  0 79 74 54  0 33 74  0\n",
            " 79  0  0 54  0 74  0 74  8  0 74 79  0  0  0 79 74 33 30 79  0  0 74 74\n",
            " 74 79 74 30 33 92  0 79 79  0 79  0 74 33 79  0 74 79  0  0 74  0 13 74\n",
            "  0  0 79  0 60 33  8  0 33  0 33 79 92 79 79  0  0  0  0 74  0  0 79 33\n",
            " 79  0  0 74 92 79  0 79  0  0  0  0 79  0  0  0 60 79 79  0 74 74  0  0\n",
            "  0 33  0  0  0 33  0  0 79  0  0 74 33  0 33 74 33 33 74  0 74 33  0  0\n",
            "  0 33 79  0  0  0  0  0  0  0  0 74 74  0 74  0 33 33  0 79 33 79 33 33\n",
            " 74 33  0 79  0 79 79  0 79 79 79  0  0 79  0 79  0 79  0  0  0  8 48 74\n",
            " 74  0  0 79  0 33  0  0 79 79 30 33 79 79 54 79 33 33 33 79 74 79 74 79\n",
            " 79  0  0 79 79 79  0  0  0 74 79 54 79  0  0 79 33 74  0 79  0 79  0  0\n",
            " 79 33 79 74 74  0 33 33 60  0 79  0  0  0  0  0  0  0  0  0 79  0 74 79\n",
            "  0 79  0  0 79 74 74  0 79 79 33 54 79 74 74 74 74  0 33 74 74  0  0 74\n",
            "  0  0  0 79 79 33  0  0 74  0  0 79 74 35 33  0 79 33 79 35 74 74 79  0\n",
            " 74 33 79 33 74  0 79 33 79  0 30  0 79  0 79 54 74  0 74  0 74 13 74 13\n",
            " 33 74 26 74  0 33  0  0 74  8 33 74 74 79 79  0 33 54  0 33  0 79  0  0\n",
            " 74 79  0 26  0 74  0  0  0  0  8  0 33 33 79]\n",
            "iter  400  loss  0.49792173504829407\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [ 0 33 54 33 33 79 74  0  0  0  0  0  0 30 54 74 33 54 54 79 33  0  0  0\n",
            " 30 33  0 33 79 79  0 33 54  0 33  0 54 33 92 33 74 33 79 79  0 54 54 74\n",
            " 33  0  0 33  0 79 92 33  0 54 59 74 79 74 79 54 79 54 74 54 18 13 33 74\n",
            " 54  0 33  8 33 33 92 74  0 59 79 79 33 33  0 74 54 33 33  0 30 92 59 33\n",
            "  0 54  0 74 33 18 33 54  0 79 33 33 79  0 54 54 92 54  0 33 13 74 54 33\n",
            " 18 74 33  8  0  0 54  8 33 79  0 33 33 79 79 54 33 54 33 74  0 54 33 79\n",
            " 33  0 54 33 33 79 33 33 79 59 79 35  0 54 35 59 79 54 74 33 33 33 54 79\n",
            " 79  0 30 35 33  0 33 54  0 54 59 33 74 54 33  0 33 33 74 79 33  0 33 74\n",
            " 54 54 79 54 33 54  0  0 33 79 54 54 54 33 54 79 74 33 33  0 79  0 54 35\n",
            " 54 33 74  0 74 79 54 33 74 79 92 79 74 54 54 54 74  0 33 33 74 54 33  0\n",
            " 54 74 74  0  0 33 33 92 92 54 33 33 54 33  0 74 33 59 54 33 54  0 33 79\n",
            " 33  0 54 33 33 54 33  0 79 33 30 92 33 54 33  0 54  0  0 74 74 33 33 54\n",
            " 74 35 54  0  0 92 79 33 74 79  0 33 54 79 54 74 54 74  0 33  0  0 33 33\n",
            "  0 54 74  0 54 33  0  0 74 74 74 33 79 33 79 79 33 79 74  0 79 54 79 18\n",
            " 54 33 92  0  0  0 18 74 33 74 33 79 54  0 74 54 54  0 54 54 33 18  0  0\n",
            "  0 92 54 54 79 33 79  0 79 74 92 54 54 33 33 54 18  0 79 54 35 33 59 13\n",
            " 35 59 33 33 35 74 33 33 74 74 79 54  0  0 33 54 74 33 74 54 79 54 59 59\n",
            " 54 54 92 33 79 35 54 74 33 79 33 33 79 54  0 54 54 33 74  0  0 13 74 74\n",
            " 74 54 33 54 54 33 33 33 33  0 54 54 33 74 33 79 54 54 54 54 74 33 74 33\n",
            " 33 33 33  0 54 79  0 33 33  0 79 54 33 33 54 79  0  0 33 54 33 33 92 54\n",
            " 54 54 54 33 54 33 54  0 74 74 33 92 33 54 60 54 79 33 74 74 74 92 18 79\n",
            " 33 74 54 33 33 33 30 33 74 74 33 79 33 74 79 54 35 74 33 54 54 74 74 33\n",
            " 33 74 33 92  0 79 33 54 33 79 33  0 79  0  0 33 79 74 74 33 33 35 54  0\n",
            " 79 54 79 35 33  0 79  0 33 33 79 33  8 54 59 33  0  0 74 54  0 33 33  0\n",
            " 79 74 13 74 33 79 74 54 54 54  0 54 54 54  0 54 92 33 33 33 79 33 79 33\n",
            " 13 33 74 74 54 33 33 33 33 74 92 74 79 74 74 79 74 79 33 33 79 18  0 33\n",
            " 74 33 18  0 33 33 54  0 35 74 54 33 79 79 92 79 79 54 74 92 79 54 74 33\n",
            " 79 79  0 92 33 74 79 74 92  0 74 33  0  0 33 33 74 54 54 54 33 35 74 74\n",
            " 74 33 74 13 54 18 79 79 54 33 33 30 74 54 33 79 74 54  0 33 35  0  8 74\n",
            " 79 74 33 33 13 54  8 33 33 79 54 54 18 54 33 79 79 79 79 74 33 74 54 54\n",
            " 33 60 74 74 18 33 79 33 74 60 79 79 33  0 79 79 54 54 33 33  0 74 74  0\n",
            " 33 54 35 79  0 54 79  0 54  0 33 59 54 74 54 74 54 54 35 33 74 54 79 33\n",
            " 33 54 33 33  0 59 33  0 33 59 79 74 74  0 35 79 54 33  0 54 54 33 54 54\n",
            " 74 54  0 33 79 54 33 33 33 60 79 59 33 79 33 33 60 33 79 33 33 92  8 74\n",
            " 74 79  0 33 33 54 79 60 54 33 92 54 33 33 92 33 13 54 92 54 74 33 74 54\n",
            " 33 79 79 54 33 54 79 33  0 35 54 92 33 33 79 33 54 35 33 33  0 33  0  0\n",
            " 33 54 33  0 59 33 54 54 30 33 33  0 35 33 79 33 33 79 79 79 33 74 35 33\n",
            " 33 54 33 33 54 74 74 33 33 54 54 92 33  0  0 74 74  0 54 74 35 33 33  0\n",
            " 33  0  0 33 33 54 74 33 74  0  0 33  0 74 54 79 33 54 33 35 35 74 33 79\n",
            " 74 54 33 92 74 79 33 54 54 33 13 79 54 79 33 18 74 33 74 79 74 92  0 13\n",
            " 54 74  8 35  0 54 33 79 74 18 33 74 74 33 33  0 54 92 33 54 33 33 30 33\n",
            " 74 33  0  8 54  0  0 33 79 33  8 79 54 54 33]\n",
            "iter  500  loss  0.35971829295158386\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [74 33 33 33 79 33  0 74  0  0  0 74 74 13 54 74 33 54 54 79 33 35 74  0\n",
            "  0 33 74 33 79 79  0 33 54  0 79  0 33 79 13 79 74 79 79 79 35 33 33 35\n",
            " 33  0 74 33  0 79 92 33 74 33 17 74 79 74 79 54 79 13 74 54 18 13 79  0\n",
            " 54  0 33  8 33 33 18 74  0 17 74 33 33 79 74 74 33 33 33 74 13  8 17 33\n",
            "  0 33 74 74 54  8 79 54  0 79 79 33 33 74 33 33 92 30 74 33 18 74 54 79\n",
            " 18 35 33  8  0  0 54  8 54  0  0 33 79 79 79 54 79 54 79  0  0 33 33 60\n",
            " 33 74 54 79 33  0 33 79  0 74 79 17  0 54 35 17 33 33 74 33 33 79 33 79\n",
            " 60  0 13 17 33 74 79 54 74 30 17 79 74 33 33  0 33 33 74 79 33 74 79 74\n",
            " 54 54 79 33 33 13 74  0 33 79 33 54 54 33  8 33 35 33 79  0  0  0 30 35\n",
            " 33 79 74  0 74 79  8 33 74 79 92 79 74 33 33 33 74  0 33 33 74 33 79 74\n",
            " 54 35 74  0  0 33 33  8 18 54 33 33 33 79 74 35 33 17 54 33 54  0 33 79\n",
            " 33  0 33 79 33 54 33  0 79 79 30 92 79 54 33 30 54 74 74 74 74 33 33 33\n",
            " 35 35 33  0  0 92  0 33 74 79 74 79 33  0 54 74 33 74  0 79  0 79 33 33\n",
            " 74 33 74 74 33 79  0 74 74 74 35 33 79 79 79 79 33 35 35  0 79 33  0  8\n",
            " 54 33 54  0 79  0 18 74 79 74 33 79 13  0 74 54 54  0 54 13 33 18  0  0\n",
            "  0 92 54 33 79 33 79 79 79 35 54 33 54 60 79 54  8 74 79 33 35 79 35 13\n",
            " 17 17 79 33 35 35 33 79 74 74  0 33 74 79 33 54 35 33 74 54 79 54 35 74\n",
            " 54 54 54 33  0 35 54 74 33 79 33 33 79 54  0 54 33 33  0  0  0 13 74 74\n",
            "  0 13 33 33 54 33 54 33 33 79 33 33 30 74 33 33 33 54 33 54 74 79 17 33\n",
            " 79 33 79 74 54 79  0 30 33 74 33 13 33 30 30 79  0  0 33 54 33 33  8 33\n",
            " 33 54 33 33 54 33 54  0 35 74 33  8 33 33 60 54 79 33 74 35 35 92 18 79\n",
            " 79 74 54 33 33 33 30 33 35 74 33 33 33 74 79 33 35 74 33 33 33 74 74 54\n",
            " 79 74 79 54 74  0 79 54 33  0 33  0 79 74  0 79  0 35 74 33 79 35 54 74\n",
            "  0 54 33 74 79  0 79  0 33 60 79 79 13 54 17 79  0 79 74 33  0 33 79 79\n",
            " 79 35 13 74 33 79  0 54 54 54  0 54 54 54 74 33 92 79 79 30 79 33 79 33\n",
            " 13 33 79 35 54 79 74 74 33 74 92 74 33 74 74 79 74  0 33 79  0 18 74 33\n",
            " 35 33 18  0 33 54 54 74 17 35 54 33 79 33 54  0 79 33 35  8 79 54 74 79\n",
            " 79 79  0  8 33 74 79 74  8  0 74 33  0  0 33 33 74 13 13 33 74 17 74 35\n",
            " 74 60 74 13 33 92 79 74 33 33 33 30 74 33  0 79 74 54  0 33 74  0 92 74\n",
            " 33 17 33 79 13 54  8 33 74 33 54 33 18 54 33 79 33 79  0 74 79 35 33 54\n",
            " 33 30 74 74 18 54  0 33 74 30 79 79 33  0  0 33 13 54 79 33 74 74 17  0\n",
            " 79 54 35 79 74 54  0  0 33  0 79 17 33 74 54 35 54 54 35 79 74 92 79 79\n",
            " 33 54 79 79  0 17 33  0 33 17 79 74 74  0 35 79 54 33 35 33 33 33 33 54\n",
            " 74 54  0 33  0 33 33 33 33  0  0 17 79 79 33 79 30 79 79 33 79  8  8 74\n",
            "  0  0 79 33 33 54 33 30 54 79  8 33 33 79 92 33 13 33 54 33 74 13 74 33\n",
            " 33 33 79 33 79 33  0 33 35 35 33 92 33 79 79 33 54 35 60 60  0 33  0  0\n",
            " 33 13 33 74 35 33 54 13 79 33 33 79 17 33 79 33 79 74 79 33 33 30 35 33\n",
            " 60 33 33 33 33 74 74 33 33 33 33  8 33 74 74 74 35  0 33 74 35 79 79  0\n",
            " 33  0  0 33 33 33 30 33 74 74  0 79  0 35 33 79 30 54 79 35 74 74 79  0\n",
            " 74 54 30 54 74  0 33 33 33 33  8  0 30  0 79 92 74 33 74 79 74  8 74  8\n",
            " 33 35  8 74  0 54 33  0 74  8 79 74 74  0 33 74 33 92 33 54 30 33 13 33\n",
            "  0  0  0  8 33 74 74 33 79 79  8 79 54 13 33]\n",
            "iter  600  loss  0.36043980717658997\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [74 74 79 33 79 33 74 74  0 74  0 79 74 30 54 35 79 33 33  0 79  0 74  0\n",
            " 30 33 74 79 79 79 74 33 30 74 33  0 33 79 13  0 74 79  0 79  0 33 33  0\n",
            " 33  0 74 33 74 79 54 33 74 33 35 74 74 35 74 33 33 13 74 92 18  4 79 74\n",
            " 54 74 79 13 33 33  8 74 74  0 35 79 33 33  0 74 33 33 30  0 13  8  0 33\n",
            " 74 33 74 85 33 13 79 54 74 79 74 33 74 74 79 54 92 54 35 33  8  0 54 79\n",
            " 92 85 33 13 74 74 54 13 13  0  0 33 79  0 79 33 79 33 74 74  0 33 35 74\n",
            " 33 74 33 33 33  0 33 79  0 35 33  0  0 30 35  0 79 33 74 33 30 33 79  0\n",
            " 30  0 13  0 33 74 79 33 74 30 35 33 35 33 79  0 33 33 35 33 33 35  0 35\n",
            " 54 30 79 33 33 30 74  0 33 79 33 33 33 33  8 33 35 33 79  0 79  0 30 35\n",
            " 60 33 74 74 35 74 13 33 35 79 54 79 35 33 54 33 74 74 33 79 79 79  0 74\n",
            " 33 35 74 74 74 60 33 13 18 33 33 33 33 79 74 85 33 74 33 33 33 74 33 74\n",
            " 33 74 30  0 33 33 33  0 74 79 30 54 74 33 33 30 33 74 74 74 35 79 33 33\n",
            "  0 35 79 74  0 13  0 33 74  0 35  0 30 74 33 74 33 35  0 79 35 79 33 33\n",
            " 35 33 35 35 33  0 74  0 35 35 35 33 33  0 60 79 33  0 35 74 74 33  0 13\n",
            " 30 33 54 74 74  0 13 35 35 74 54 79 30 74 35 33 30 74 33 13 33 92 74 35\n",
            " 74 92 33 30 79 33 79  0 79 35 33 33 33 60  0 54 13 35 74 33 74 60 35 13\n",
            " 74 35 79 54 35 35 33 79 35 74 74 33 74 60 33 33 35 30 35 33  0 33 85 35\n",
            " 33 33 33 33 74 74 33 35 79 79 33 33 79 13 74 92 54 33 79 74 74 13 74 74\n",
            " 35 13 30 60 33 33 54 33 60 79 33 79 60 74 33 33 79 33 33 33 35 33  0 33\n",
            "  0 33 79  0 33 74 74 30 33 74 33 13 33 30 30 79 33 74 33 30 33 33 13 33\n",
            " 79 33 33 33 54 33 54 74 85 74 54 13 54 33 60 54 79 33 74 74 35 54 92 79\n",
            " 74 35 54 60 79 33 54 33 35 74 33 74 79 35 33 33 35 74 33 30 18 74 35 13\n",
            "  0 35 60 33 74 74  0 54 33  0 79 74 74 74 74 79  0 35 74 33  0 35 30  0\n",
            "  0 30 33 35 79 74 79 74 33 60 79 79 13 30 35 33 74 74 74 33  0 33 79 74\n",
            "  0 35 13 35 33 79 74 54 33 30 74 54 54 33 74 33 54 79 74 13 74 33 79 33\n",
            " 30 33  0 35 54 79 74 35 33 35 54 74 79 35 35 79 74 79 33 79 79 54  0 60\n",
            " 35 33 18 74 33 54 30 35  0  0 54 33 74 33 33 79  0 33 85 13  0 54 35 79\n",
            " 74 79 74  8 33 35 74 35  8  0 35 33  0 74 33 30 74 30 30 33 35  0 35 35\n",
            " 35 30 35 30 79 92 79 35 33 79 79 30 35 33 60 79 35 30 74 33 74 74 13 74\n",
            " 74  0 33 79 13 54  4 33 35 33 54 33 13 33 33 74 74 74 74 35 79 74 33 33\n",
            " 33 30 74 35 54 33  0 33 74 30 33 79 33 74  0 33 13 33 79 33 35 35  0  0\n",
            "  0 33 74 79 74 33 74 74 33 79 79 35 79 74 33 35 33 33 85 79 74 54 74  0\n",
            " 33 30 79 33 74 74 54 74 33  0  0 74 74  0 35 79 54 79  0 33 33 33 30 33\n",
            " 74 54 79 33  0 33 79 33 79 74 35  0 74 79 33 74 30 74 74 60 74  8 13 74\n",
            " 74 79 79 79 33 30 74 30 30 79 13 30 79 74 18 33 13 33 33 33 35 13 35 33\n",
            " 79 79 33 33 33 79 74 33 74 85 30 54 33 79 79 33 54 35 60 60 74 33 74 74\n",
            " 33 13 33 74 35 33 13 30 30 33 33 79 74 79  0 33 74 35 79 33 33 30 85 33\n",
            " 79 33 33 79 33 35 74 33 33 33  0 13 33 74 74 35 35 74 79 74 35 79 79 74\n",
            " 33 74 74 30 33 30 30 33 74 74 74 33 74 85 79 33 30 33 74 35 74 35 74 74\n",
            " 35 33 30 60 35  0 33 79 54 33 13 74 30  0 79 30 35 33 35 79 35  8 35  4\n",
            " 79 35 13 74 74 54 79 74 74  8 79 74 74 74 33 35 79 54 74 33 30 33 30  0\n",
            " 60 79 74  8 33 35 74 33 79 33 13 79 33 30 33]\n",
            "iter  700  loss  0.3451830744743347\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV8AAAD4CAYAAABL2+VjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydd3hUxdeA39mWQkJL6L0jVSSAoAhSFCWCFCkqIkqxoKKiYv/ZO2LhQxFQVJCOFJEivUgHkU7oLSQBQnqyZb4/dhMTkmy9ySZh3ue5T/beO+Uk2T0798wpQkqJQqFQKAoXnb8FUCgUihsRpXwVCoXCDyjlq1AoFH5AKV+FQqHwA0r5KhQKhR8w+FuA7ISHh8vatWv7WwyFQlEM2LVrV5yUsoIvY9QXQqa42fYirJBS9vBlvuwUKeVbu3Ztdu7c6W8xFApFMUAIcdrXMVKAUW62/R+E+zpfdoqU8lUoFIrCROA/JaiZzVcIoRdC7BFCLHWc1xFCbBNCRAkhZgshTFrNpVAoFFqgA4LcPApibq14DjiU7fwT4EspZX3gKvC4hnMpFAqFzwjA6OahNZooXyFEdaAnMMVxLoAuwDxHk+nA/VrMpVAoFFqRaXZw59AarcacALwMhDrOw4B4KaXFcX4OqJZXRyHESGAkQM2aNTUSR1FYXDlxgnN//03ZWrWo3KoVplKl/C2SQuE2mStff+Cz8hVCRAIxUspdQojOnvaXUk4GJgNERESoLD/FhPXvvsv6995DWiw5rldr356B8+cTWqWKnyRTKNzHnxtuWsx7G9BLCHEvEAiUBr4CygohDI7Vb3XgvAZzKYoAU269lfPbtuV57/zff/Ndq1aMvXABoVMxPIqijZYrXyHEc8AIx7A/SCknOGvv86dDSvmqlLK6lLI2MAhYI6V8CFgL9Hc0Gwos8nUuhf85OH9+voo3k5RLl9j29deFJJFC4T1aeTsIIZphV7xtgZZApBCivqu5C4pXgBeEEFHYbcBTC3AuRSGxbPRot9pt/uSTApZEofAdDb0dbgK2SSlTHE/764G+zjpoau6QUq4D1jlen8D+LaAoIZhTU0mOjnarbXJsbAFLo1BogwdKMFwIkT0Ed7JjzwpgP/CBECIMSAXuBZyG66oIN4VbJMfE8H3r1m63F3p9AUqjUGiDhzbfOCllRF43pJSHhBCfACuBZGAvYHU2mFK+CpdcOX6cbxo1AqvT91IOGt53XwFKpFBog5beDlLKqTjMq0KID7G72OaLUr4Kp5zasIHpnTp51kkIev3wQ8EIpFBoSOaGmxYIISpKKWOEEDWx23tvddZeKV9FDqSUxB0+TNyRIywdMYKUuDiP+gu9nuHbthFUrlwBSahQaIfGQRbzHTZfM/C0lDLeWWOlfBVZnN64kVm9e5MWHw8eVrU2lS5Ny6FD6frBBwSEhrruoFAUATQ2O3T0pL1SvgqklGz7+mtWjBnjVf+2zzzDPcqvV1EMKdbhxYrijZSS+YMHc2DOHK/6C52OHl99pbFUCkXhUNzDixXFkKsnT3Jx926SoqM5uHChx2aGTHQmE2lXrxJUvrzGEioUBY9a+SoKDZvVyu+PPsqhefPQmUxkJCZ6rXgBrGlprBw7lt7TpmkopUJROAgKJlG6O6jMJzcY277+msMLFmBJSyMjIcEnxZvJwblzNZBMoSh8BGA0uHdojVr53mBs/+YbzCnu1mt1D5UHVFFcEQIM7mpBi+smnqCUrx+4sGcP2ydMID0xkVtGjKDBPfcUyrxWs5mEc06Dbrzipn79NB9ToSgMhACjnyLhlfItRLZ8/jmrXn45x6P+4YULMZUpw1MHDlCmWp7FPjRj00cfYTObNR0zKCyMyEmTNB1ToSgsPFr5aoyy+RYSW7/6ilUvvZSnjTXj2jUmVK/O+R07Cmz+Va+8wrq339ZuQCFo2Ls3L0ZHYwzy15aFQuEbQoAxwL1Da5TyLQRsVqtbAQxT2rbl4t69ms9/Ys0atnz6qU9jZM9SZgwOpnLLljwwaxZ6fy0bFAot8GMFTfXJKQR2eZBk5oc2bRgXH69pIcq1b77pU39jcDBdPvyQE6tWkRYfT9OBA7ll+HAMgYEaSahQ+Ak/RlloUUAzENgABDjGmyelfFsIUQeYhb2KxS5giJQyw9f5iiNRy5e73VZaLPx2330MXbNGs/kT3dxkCwoLo+5dd3F+61aSLl5EbzIhpaTnxIm0GDKEW597TjOZFIoiQ3FVvkA60EVKmSSEMAKbhBB/Ai8AX0opZwkhvgMeB27InZmQypU9an9q7VreCwhAZzBQt2tX7vriC8IaNPB6/vo9erBr8uR87zeIjCRy0iRKV6+ede3ysWOkXb1KpZYtMQQUgMFLoSgKCMBP3g5aFNCUUsokx2lmuSMJdAHmOa5PB+73da7iStunnvK4jy0jA0tKCkeXLmVK27YkXrwIgDklhYRz50i6dIkDc+ZwZMkSLGlpTse64623MOSxKaYPCGBsTAwPLlmSQ/EChDVoQLW2bZXiVZRsNLT5CiGeF0IcEELsF0L85rAK5IsmC24hhB67aaE+MBE4DsQ7CsmBPaN7nn5UQoiRwEiAmjVraiFOkaNSixaUrlmThDNnPO8sJWnx8ax75x10ej17f/wRm8WCzWxGHxCA3mRC6HQMXrKEWh3zzmhXulo1Rh85wrKnnuLkmjUIvZ7GffrQc+JETCEhPv52CkUxRmA3mPo6jBDVgGeBJlLKVCHEHOzV3H/Kr48myldKaQVuFkKUBRYCjT3oOxmYDBAREVFig6VG7d7NxCZNSImJ8ar/7u+/Rx8YiDXbKteano41PR2AmT17MjY6GmNwcJ79y9SoweAlS7yaW6EosWi74WYAgoQQZiAYuOCssaauZo7M7WuB9kBZIUTmr1UdOK/lXMWN4LAwXrxwgb4zZ9qdC73A6sS8kJGYyJc1a3IlKspbERWKGw/PzA7hQoid2Y6RmcNIKc8DnwNngIvANSnlSmdT+6x8hRAVHCtehBBBQHfgEHYl3N/RbCiwyNe5ijs6vZ7mgwfzYnQ0dbp391oJ50fq5ct806ABl48d03RchaJEo3fzcFQvznZk7WILIcoBvYE6QFWglBDiYWfTarHyrQKsFULsA3YAq6SUS4FXgBeEEFHY3c2majBXiSCkYkUeWbmSNzMyqNaunebjz+7bV/MxFYoSiXYbbt2Ak1LKWCmlGVgAdHDWwWdrh5RyH9Aqj+sngLa+jl+S0RkMPLZpE7/ecw8n//pLs3FjDxzQbCyFokSjnc33DHCrECIYSAW6AjuddVDhxX5GZzDQ8pFHEBqG6epUyK9C4R6Z3g7uHE6QUm7D7lq7G/gXu27N37keFV5cJKjUogV6oxGLRZuEoS0edmpqUigUmWjo7SClfBtwO3uVWvkWASq3bEnN227zePUrDIZcm3YVmjVTBS0VCndRiXUUg5cs4buWLbl89KjbfRpFRjJgwQKiVqzg8rFjVIuIoPqttyI09qJQKEosxTm8WKENhsBAGvfpg95kcqu9LiCA2nfeiRCCBj16cOszz1CjfXuleBUKT/Djylcp3yJExJNPojO6V8jaYDQq265C4Ssabbh5g1K+RYiytWrx0LJllK1dG0NgIHqTiZAqVUCf+7nIFBKivBoUCl9RNl9FJrXuuINnT5wg8fx5DEFBmFNS+LpePWxWa452adeusXvKFNq/8IKfJFUoSgB+TKauVr5FECEEpatXR28y8WOnTnkWvbSkpnJg9mxkHjXhFAqFByibryITKSWbP/2UzypU4NrJk/m2u7BzJ1/XrUvC+Rs6Z5FC4T2Z3g7u5XbQFKV8iyC7Jk9m/TvvZKWLzA9psxF/6hRTOzgNIVcoFPmhvB0U2Vn/v/9hTklxu33CmTOc3rChACVSKEooyttBkcn57dtJio72uN8/v/xSANIoFCUc5e2gyGTRY4951e+khtWOFYobhuJcOl6hHWkJCV6ng7x25gwZycmYSpXSWCqFogSjXM0UAFs++8zrvkII5HW+wIqixcaNV7jlls2UKrWCMmVW0rPnDv79N8HfYik08HYQQjQSQuzNdiQIIcY466NFGaEaQoi1QoiDjrLJzzmulxdCrBJCHHP8LOfrXCWdfb/+6l1HIajcsiUBpUtrK5BCM14Ys4++d8xj/55oUlJsJCRYWLYsjhYtNvPkk/v9Ld6Ni0Y2XynlESnlzVLKm4HWQAr2YsL5osXK1wK8KKVsAtwKPC2EaAKMA1ZLKRsAqx3ninxIT0wk2UVlY6HXozeZMAQHY3SYFwzBwQSWKUPvn34qBCkV3vBO/y8xfXUb/fkQmbWE+i8B0nffnWXevIv+Ee5Gp2C8HboCx6WUp5010qKM0EXs1TqRUiYKIQ4B1bAXk+vsaDYdWIe9rpsC2D11KmvefJO0K1coW6cOSOnUr7fG7bfT5+efCSpfHkNAAAfmzOH89u2ENWxIiyFDCCqnHiyKIgdWbCRj/qsEkc4euiPJO+vc6NEH6d+/SiFLp/DQ5hsuhMheGmhy9iKa2RgE/OZqME1NzUKI2tjruW0DKjkUM0A0UCmfPiOBkQA1a9bUUpwiy1+vvsrmjz/OOr98+LDT9k0HDaLvr7+iy5Zgp+Ujj9DykUcKTEaFNiwY+wEGMgCwocOaz8PmpUsZhSmWIhPPlG+clDLC6XBCmIBewKuuBtNsw00IEQLMB8ZIKXPsIkh7AoI8kxBIKSdnlmKuUKGCVuIUWWxWK1s+/dTt9k0eeID+v/2WQ/EqigepqVbO7T+BDokEDtARv2XuVuSN9uHF9wC7pZSXXDXURPkKIYzYFe8MKeUCx+VLQogqjvtVAOcGzRuEmAMHkDab2+2rtmlTgNIoCpLxX5zgOC2xYOA8DYmjBuQyO9jXJAEBKgm+X9A+yGIwbpgcQBtvBwFMBQ5JKcdnu7UYGOp4PRRY5OtcJYFLe/d61H7Txx+TdMnll6iiiCGl5IN3D7OVvlwjnKO0wUJeVUrsSveZZ2oVroAKOwIIdPNwNZQQpYDuwAJXbUGble9twBCgSzYft3uBj4HuQohjQDfH+Q3Nvl9/5Y8nn/Sojzkpia1ffllAEikKim+f/D/SzZBMWb5hKht4EJnPx81kEnz4YaNCllABaGp2kFImSynDpJTX3JlaC2+HTeR+lsqkq6/jlxSklKx86SWPEuYAWDMyOLF6dQFJpSgILOnprPr+D2yMdlxxblLYtKkdRqOKd/ILKsKt5PPvjBkke5EwB6Bc7draCqMBV0+c4MfOnfm4bFm+rFWL7RMnqsTuDtb//g+naOJW26pVjbRpo9wE/YpKKVlyubBrFwuHDPG6f/uxYzWUxnf+GjeOr+vV48z69aRfu0bCmTP8OXo0s/v29bdoRYJX34/jKlVwteIFaN1aKV6/opKpl1yOLF7MlLZtvR9Ar6d6u3baCeQjfz73HJs/+STPe0d+/50rUVGFLFHRI/bgARqzhXrsJJAkp23Hjq1dOEIp8kallCyZmFNSmP/ggx65ll2Pvoj491rS05ncurXLrGuHFy+mww1c1PPCrl08bHsNiY4IlqPHwjoeYjMPkN3V3UQyAaFluOOOMP8Jq/gvvNgPKOVbgJxcu9bn4Ijad96pkTTeIW02Ds6bx/LnnnMryfuNnlltxr33IrBhwILREdnWiZlcojaXqEkS5enFBDII4Ylff/CztAqVz7eEInQ6nzah9IGB3PXFFxpK5BnR+/YxpW1bl7XkstN0wIAClKhoc2TJEpJiYtFfF8xpIp12LGIbvRjB8+ix0eTrNfTqlWfEvaIwUcq3ZFK7c2evlW9I5coM27yZ8nXraiyVc6SUICVXT55kSrt2HilegLK1btxggbVvveVInJP7f16Ka4DAiJmpugkcfbhFocunyAOlfEsmxqAgBsybx8zISKTF4lYfU2go7Z59lo6vv44xKEhTeaSUHF26lL0//khQWBi3vfIKV6Oi2D5xIpePHiU1Lo7UK1fsK3YAD23VOsON/Xa69M8+dOT+m5kxcZQ2hHEeA2aaGbZx4EBfbr+9vB+kVOTCT9sqN/anpRCof/fdPH/mDNNuu434kyfzbVexWTP6/fYbFZs101wGKSX/TJ/O0iefxJqWlnV9z5Qpebf3coPQZrFw+dgxwho08Kp/cUVKyQsvHCJQBhJIClb0CGzokJgxkURZttKHOuzFSAZ1M/6mWjU34lUVBY9a+ZZsQqtU4ZmjR5nepQsXdu7EkpqadS+wfHnu+uwzbh42DHuaDO9IuXyZg/PmEVSuHA0jIzEGBwP22m6/9e7tcU4Jbzm5bt0Np3zfeGk33084zu3cQ1uWYCCDa4STRijHackZmpBGKGmE2J8oQsKoUyfY32IrQHk73AjoDAaGrl1L1J9/cnTJEoLCw2k1bBjl69f3aVwpJQsefpj9M2dmXRMGA/1mzqRSixZMjojAnOTc11QrJLBq7FiaDxpEQGhooczpb47vPIzliztpzmBiqUYC5SlPNGWIowxxVOQUF2iIkTSasBkzAQz+UtUUKDKolW/xwZKezob33+fY0qWEVKlC148/pnIL9zZPdHo9DSMjaRgZqZk8Wz7/PIfiBZAWC/MGDCA4PLzQFC/Y38fmlBT2TJvGrc89V2jz+pM5kT0IIpHu/AgIDGTkiGsTSO7nS6bzIc1Zi+w6lm7D7/WXuIrr0VD5CiHKAlOAZtjXIo9JKf/Or72KcHOTjORkdkyaxMelS7Px/feJ3ruXqD//5PuWLdn61Vd+kyt7RYzrSYmLK0RJ7EiLhROrVhX6vIWNlJK5AwaQcek0AjBixnid4s3EgJlhvIwOG1frqhDsIoW2EW5fAcullI2BlsAhZ42V8nWD2IMHmVCzJsuefhprRu5yLytfeAGr2ewHyeyFN4sSwmCw16Qr4ax+9VUOzp3rVlt7+gCJDgsLf1aViosaUu/e4QwhRBngDuy5zZFSZkgp4531UcrXDWbdfz+pV65APj670mZj+ZgxWLJ5EhQ01owM5jzwADY3XdgKC4PJRJunnvK3GAXKlaiofPNbOEOPleR0g8r+VoSQOsgIdO/AUUAz2zEy21B1gFjgRyHEHiHEFEdy9XxRytcF+2bM4MqxYy7b7Zw0ia/q1uXa2bNO20Xv3cvKsWP585lnOL1hg9cfxHXvvMOxP/7I9wuhsJGAMSSE/nPmUOGmm/wtToGybPRo143ywIqBO5vH++TVotAWKcCi17l14Cigme3IXrnYANwCTJJStgKSgXHO5taqhts0IUSMEGJ/tmvlhRCrhBDHHD+LXe48abOx/Pnn3WwsSYqO5o8nnsi3ycYPP+SHdu34e/x4tn/7Lb/ecw9LR43ySgHv+u67HC5rnqI35VXSxnsEYE1PZ9XLL7v8AirunNu61at+Ej3j3nVa/FZRyEghsBoMbh0uOAeck1Juc5zPw66M80Wrle9PQI/rro0DVkspGwCrcfEtUBTZ8OGHpMbGut9BSqJWrMgzSGH3jz+y5vXXsWVkZK1WLSkp7Pv1V879ne+GaL5kJCd73AcgqEIFIidPptPbb2MMDiagdGn0JhM6o9Gr8bJjM5u5fOQIs3r18nmsoozBi8hDCVjK1KLD/T6kF1UUCFa93q3DGVLKaOCsECKzHlRX4KCzPpooXynlBuDKdZd7A9Mdr6cD92sxV2ERtXw5mz780ON+eT1SbvzoI5Y89lie7S2pqRxa4Fa9vRzU6tgRvHh8TY2NZdlTT5GemMjYmBiGrlvHmNOneSM9nft/+cVnJSytVi4fPcplN0w1YPcaKG420JvySBovgXM05AyNsVy3NS6BDAKYbXi3cARUuI1EYEXv1uEGzwAzhBD7gJsBpwqkIG2+laSUFx2vo4E8UzgJIUZmGrBjPVllFjCbPvnEq8f6BpGRCJ2OE6tXMzMykq/q12fNa6857ePNSqrH11/bAxm8UMA2i4W/P/+cxAsXqNKqFSGVKyOEoF63bjlSWAqDAVNoKD2/+86jeYTBQNrVq07bJMfGMnfAAN4PCOA9o5EZPXsWG3PFXZ9/nmW2MWPEjIl/6MxUvuA33uE8DckggDSCsWBkN91ZxigCwqv4WXLF9UgEFvRuHS7HknKvwxbcQkp5v5TS6YdAaLXqEELUBpZKKZs5zuOllGWz3b8qpXRq942IiJA7d+7URB5f+aZBA6+qMjz2999c2LGDv155xW3lHRQWxkN//km1Nm08mivh/HnWvf02e6ZO9VhOgLvGj6e9w6ZtSU/n24YNSbhwISsJkM5gIKRqVcwpKaR64DNsCg3lpdhYDAF5x23arFYm3nQT8SdPZnlrCJ2OUhUr8uzx41mh0UWZc1u38muPHlxMq8hv6c8TS+0c98M4RyiXuUQdUimNgQw+Hd+C558v+W54hYUQYpeU0icjeosIo1y6070ER7VEjM/zZacgV76XhBBVABw/YwpwLs2pfeedCA+zdAmDgV+6dWP5s896tGpOvXyZKW3bctzD4ITS1arRa8oU7p4wAeFF0vaD8+ZxaOFCbBYLhxcuJPXKlRzZ12wWCwlnz7qveIXAEBxMz4kT81W8YDfpJEVH53CTkzYbadeusemzz7i4Z0+RWQXbLBY2fvQRX9evz2cVKzKrTx8u7dtH9VtvZeylSzw170s63lkV3XWfJDMBNGIrQ3idvnxCJXGKu+4K988vocgXjc0OHlGQK9/PgMtSyo+FEOOA8lLKl52NURRWvrEHD3J81Sr0JhOrX3uNjKQkt9NB+orOaOS15GT0XthdL+zcydavvuJKVBRpV69y5fhx+8afiwxlxlKlCKlUCUt6Oonnz3srOgBlatdmwNy5VI1wvjjY8sUXrH71VWzOAlN0Oqq2bs2Df/xBqQoVfJLLE5IuXeKPp54iavlyLOnpkEdlDkNwMMM2bKBq69aA3W49YMBeFsyPxibtq94RjMFABgYsSMCGgQGLFtOs1z2F9ruUdLRY+TaPMMmFO91Lat9AnNN05atJVLMQ4jegM3Yn5HPA28DHwBwhxOPAaaBIlzg4+scfzBs4EHM2LwK9yUTtO+/k/NatmFNSCrxEjs1sJmrFChp5kfuhakQEfX/5Jes87do1Tq1dy7zBg3Okkbwec3IyV0+c8Ere6wlv3JjKrVq5bFexaVN0BoNz5WuzcWHHDqZ16MAzbm7e+UrU+r+Z2Lk3pYl1+khoSUlh1csvM3T1asC+yTp+fGP++P00qRYj3fgJE6k4siI7ItwsLBkyiKbxys+3KJFp8/UHWnk7DJZSVpFSGqWU1aWUU6WUl6WUXaWUDaSU3aSU13tDFBmOr1zJrPvvz6F4wR5FdnLVKjISEwutNlnMvn2ajBNYpgyNevVCX4gJzo8vX85HISFs/Ogjpx4MtTt3drtCxpWoKGb07EnCuXNaiZkn89/7iemdO1LGheLN5PyOHTnOa9QIYkyNr6jDPmrzT5bizU56QgLrfv9XI4kVWmA3OxjcOrRGRbgBy8eMKTzTgovghhq33abZXBnJyV77A3uLJS2NNa+/zqLHH8937lPr12Ms5TTyMgdRy5YxoXZtZvXrx87vviNDo0xtUkqi9+5l9f/e55+3HseANc/EOHn2Dc5tv721Sz2G8hpB5P83f2vor1itxcu1riQjEWRgcuvQmhte+a5/913iDjlNPqQpNoslX19aY3Awte64Q7O5zm3d6p/wYyn558cf+aJyZU6tW5frtjU93eNHb2m1cmTBAla88ALfNGxI4oULPom4f/ZsPgoN5ftWrdj4zpt5lv/JDxs6wu5/Ntd1d4qHVk3dwfbtTvOtKAoRCZq5mnnKDa18o/fu9SpBii+UrlqVu8ePz1XvTOh0DF27VlN7YEDp0l55QWhFRlISv/Xqhfk6z4/anTt7nQXOkppKckwMK8eO9aifzWLh8O+/M/fp13m/TX/mDxqUZWYSjsMV0nFs1T/AoI9zJw9KvXLF6YpeAEG2a5jNauVbdPCf2eGGTqb+78yZuRRDQaILCKDT229zy/DhVGzenM2ffMLV48ep1akT3T75hKBy2qa/qNa2LaUqVybJRy8GX8hITGTugAH0/fVXAsuUAexfCvd9/z1LRo2yZ4LzcHUurVaOLl3qsl3MgQOcXLMGY1AQGz/+mCunLyAs9v+3J19xEkghhF36+9ll6MO0mR0oWzb3Y2h5F+WTkghlnW0gW0fuJyhIxyuv1GXgwCpqA86PZLqa+QPNXM20oDBdzaSU9ixkp05pMp4+IMDlJlL3zz+n/QsvFOqHLe7wYX7s1ImU2Fi/ZkDTBQTw1L59hDVsmHXt8tGj7J0+nZNr13Lew/wWQWFhvJyP/3FyTAw/d+tGzP79Pv3OEoinIhsNQ6ndaxB33x1Onz6VqFAhbx9mKSVT27fn4p499hwe2UglmB+YwBWqkan6dZjpdpuJFZuuT4uicActXM0aR5SSk3e6V7S2k9hebIIsigxWs5m906czrWNHvm/Vihn33ssPbdtqo3iFoMNLL/HixYv5BmboTSYGL11KhxdfLPRVTnjjxoy9eJFhGzYwaNEiKjRt6vEYYQ0behUCnR1bejozevbMNW7XDz5g+JYt9Js926PxGkZGknbtWpZXhZSSQ4eS2LXrGjP79ifm33+9Urx2n1wdFgwsZyRb2y3k42UvM3/+LYwcWTNfxQt2l7OHV6zgpr79sGLAhiCdQK5Skd95nqtUIfua24aR1ZvT+f5xz3OIKLShRARZaEFBrHytZjOTmjfn8pEjmo6rMxio2akTD8yeTXBYGGAv2zMzMtL+wdfpsKalUadbN/r8/HOhBgo449y2bUxt3z6XYipTqxbJsbFYUlKyrgm9nmaDB9P3l1+IO3KE2AMHOLxoEceWLrUnl8+G0OlAp3PpNXJT//4MuK4ChJSS/b/9xsqxY0m6eDGfnrkRBgMhFSvSaOwEXvy+EmfPpoE1A1t6Co3ZwjHakEooVThODyZTg8Mux5SAsfMTtHp2LN0j62A0erc++X7SCV568QCJqfYv47JcJJ7/cjtU5ji38jvluUAgybyxayFVbnGagVBxHVqsfBtGhMqvd7o3xD1inaYr3xKvfFe/8QabPvhAs/GaP/QQ90+fjs7JRlbc4cMkXrhA5ZtvJqi8e3HjhSpiYcMAACAASURBVMnJNWv4fdgwEs6eRW800vyhh4j84Qc2ffABmz/9FCEENouFW0aMyHNzEOD0hg0sePhhUi9fxma1UqllS2refjtbx493LYAQlK5enYhRo4g9fJjjK1aQevVqLsVtLFXKbpN3EqVnRccE8TNJlEPK7E8VkuyrTCNpPM4LVOZU1l2JQCCzWtW8/XbumzqV8GymEV+YN+8i77wTxbXTp6mYuJ3d3IVETyO20o9PMGBGhw0bgsDQEEYfOUJoFZV8x120UL4NIkrLCTvdS/MZKVYr5esO0mZj36+/8vvQoZqMZyxVipE7dxLeuLEm4xVVzKmpJJw7R2iVKphCQpy2lVJy9cQJjEFBhFatyqV//+W7Vq3yDMktKKK4hTm8Sgau/IZt3MQWBjqy/FnRs8L4DCOfrMcDL/chpHJlp1+onpKemMg/06dzfOVKjq9cyYX0KvzBk3ThZ2pxINeGn9DraT1qFD0nTtRMhpKOFsq3fkQZOX7nrW617S1WOp1PCHEKSASsgMWVbCXS2+Hinj3M6NmTZA8eYfMisGxZwho1ot5dd9F29GhKVayokYRFF2NQEGEudu0zEUJQvl69rPNKzZsTMWoUO//v/wpKvFwkUzbb2tUZOqKpa7fpCgOlRv3Egjf6Ua1aoOYyJV68yA8REaTFx2N2mHEqcYpHeTVff2JptbLr++/ZNXkyZWrU4M5336XFww9rLpsiNxr78N4ppXQrE1WJU76WtDR+7tKFtHjvHdl1RiPdP/2Uds8+a7dlKtym58SJNLj3XpaOGuVzoh53qMFBpFsfHhvlucBpQ2tGzx9P617aBbNcz5rXXyc5JiZXcVNXgRw2qz3CLv7kSX4fNoz0hIQSX4zU32SGF/uDEqFZLv37L/MGD2ZikyZ806iR14pXZzLR7dNPeeXKFW4dM0YpXi9p2LMnL5w7Rzk3V9C+UJ5obmYVRv7z1xZYIJei01Hvrrt54/D6AlW8AEcWL/aqqnQOi7XFwrKnn+bAH39pJ5giFx56OzirXmwfDlYKIXblcS8XxXrla05NZertt3Np926fx2oxZAh3ffFFkfFKKAkMWb6ciU2auJ1Ex1t68n/UZj/b6YmZIJqygVRKsZNI0gmiUYMgJk5qQdeuhZNP1xDovikj57ZgbmZH3k3Is3N4+at+PsulyI1EkO5+3oY4F3bc26WU54UQFYFVQojDjhJreVJsle+lffv48Y47SL92zadxKt18M6N271ZRRgVAubp1GbxkCXP69SMjMdFpW2Ew0LR/fxKiozmTRz4Ip32BZmygGTnf592YTsvHhtNn6g8eSu4brUeNYtOHH9qj95zgzla3DhvJX/dnWvwrPDb9Y20ELAJIKUk8fx5jcLBfPYK0NDtIKc87fsYIIRYCbYF8lW+xe662WSz83LUr37Vs6ZPi1QUEcPtrrzFyxw6leAuQet27U6drV5ftpNXK1ZMn2fH3ZXaZ+hEtGoDOt42Q0lWr0mtS4XsP3P7KK9Tt3p1c5S2uw52cEgL7h/TMz59xYs8xjh1LLtZZ0Y6vWsUPbdvyfkAAX9aowadhYbwjBJ+UL89GN76wCgItgiyEEKWEEKGZr4G7gP3O+hS7le+ixx7j5Jo1Xvc3BAbSf+5cGvbsqZRuIRFSubLLNlYJn227m5PcjA0desyUkTE8xdMe5WHIpGGvXjwwe3ZWocvCRG8yMXjxYta99x7r33rLrT6uzA9g46E2X/FvYC+CgnRMnNiUAQOKj0/wsWXLWDBkCGlX8k7rnXb1Kmtef50NH37Io2vXelzP0Fs0zO1QCVjo0CkGYKaUcrmzDgW+8hVC9BBCHBFCRDnKCXmNtNnYN2OGV30Dy5blkdWrGZeQQKPISKV4C5FGvXq5bLOHuzjJzZgJxIqJDEoRSy0u4NmmnT4ggNvGjWPwokUe2V615srx42z52DMzgav1bIZVT3Kylbg4Mw8+uJc2bTbzzTenSE4unFzUnpCRlMSRxYs5vHgx3zZpwsyePfNVvNmxJCfzS7duXm1YeoNW4cVSyhNSypaOo6mU0mVkV4GufIUQemAi0B04B+wQQiyWUh70ZrzECxdc1iTLC2OpUjyyerUK3/QTdbt1wxQS4jQJ+g4iMWNXlgKrw31Mxxxe4xlGYMD1h9EYHMwdb73FbS+9pJXoXrNr8mSP0mYK/ktZmXmeHYngMO2zzq1W2LkzgYMHk5g06QzbtrUnJMRQJBYVB+bOZdGjj2IDrNnC1d0lPSGBE6tXU//uu7UX7jrsG2755+soSAra7NAWiJJSngAQQswCegNeKd/AsmVdN7qOqm3bMnjRIrcefRUFg95o5L4pU5j/4IP5fnnaMNCEDdzFNMoQQwqlWc9g9tCdeCoSTv7J03UGA8O3b6eKG/XjCouEM2ec16jLgyy1KXRIactSyADLGUEKZXL1SUmxEXM4itGln6Gubj91unWj/8yZWflGCgqbzcbfX3zBzpm/s/VSE6S00SD0DE3a1SFq9k8e/+7Xc3bLlkJTvv5KKVnQyrcakL0G+DmgXfYGDn+4kQA1a9Z0OpgpJISg8HC3S5mHNWrEiG3bPJFXUUA0GzgQncHAvIED86yH15lfacAOTNhTMZYigW78RDDXCHOieAH0gYFU9CJbW0FS9667OLxoERYv8kULnaDFw0OJPXCAUlVr8NLynlzMqAqADistWM0trEACcVSnhVyDASvYIGrlX4wM70H5Tn0Z9tlwWrXRxnXSajaz4KGHOLRgQdb/L/OLoRFbkIAuGg4ds9syfV1/h2mUX8MdSqrydYmUcjIwGey5HVy1v/+nn/jtvvtcpgtsdP/9DJw/XxshFZrQpF8/+s2cye+PPppLKTVhc64PrIl0OjLH6QdZFxBA5//9zy8ba85oPngwmz/9lMuHXWdSux5ptWJNTWWko0jnsf8d4/PPT5KcbGEIr1GL/VkFOmtyKGuFnE4AC3iZk7RErtcxqe0mGun38PmLZloMfABTcDBhjRqRkZTE+e3bOfrHH1iSk6nZsSNN+vd3aiP/7uabiTuY84FV5PFaCzUm9Hoa3XefBiO5pthXL3bCeaBGtvPqjmte07BnT/rNnIk+ILedJrhSJR6YN483LRYGLVyoItSKIE0HDKDN00/nup6fgtW7CMkNrVKFDi++qIFk2mIIDGTE9u3cMmKE/X3osMUaAgMJqVrVZf+k6Ois12+/XZ+pU5vTPvCvHIoX/vu7CcBIBv34lJtZhZlALARy1NqKrz/dzw+tW/N/LVvyvsnEx6VL80u3bmz78kt2TZ7Mwkce4as6dUjIJxx81+TJuRRvQRL53XcElC5dKHP5s3pxQa98dwANhBB1sCvdQcCDvg7abNAgmg4YwJnNm7m4Zw9B5ctTr1s3ZdctJoRWrYohMNBtn05nbljJ2ZRUUSMgNJT7Jk+m64cf8s/PPxN/8iQ177iDxvffz+w+fTj2xx959tMZjTTMtvITQtC/TxiHB33p1CNCh8REOt35kX10JY0QzASyk560YykyIyPv/lKSFB3NsqefZtDvv+e4ZbNYWPHCC57/8l7SbswYbhk+vNDmgxJqdpBSWoQQo4EV2J9IpkkpD2gxttDpqNWxI7U6dtRiOEUh0vzBB1n7xhtut3dmdggpBvlvg8PDaX+dAus5aRKTIyJIiYnJcV3odJSpVYvWo0bluo4QCDdSwFoxUINDHMPuK2sh72rZ13N06VKklDk8Ji7t2+d1sVNPCAoP555vvqH5oEEFPld2MkvH+4MCt/lKKZcBywp6HkXxIaRSJQYtXsy8gQPJSEnB6m0RUyG48913tRWukChTowZjTp3iwOzZ7J0+nfiTJwkoXZoWDz9MxBNPEBAamqO9zmCgVqdOnHYj9FogSSMYAD3pNM0/wjVnv+vc1GIOHmThI4/kqkenFabQUO6bMoXGvXtjyMOMWBj40+br9w03xY1J3a5deWDuXGb08L54ZOVWrYp1zltjUBA3P/ooNz/6qFvtB8ydy4TatbNK3ueFDUEawZzjJkykUJo4bmduvu2zU79HjywFvP3bb/nzmWfc6ucKodMRVL48t4wahbRa0en13NS3b5Hwu/dnSkmlfBV+Y/W4cVi9XFVZhYnmD/q8fVCsCA4PZ1x8PCvGjmX/jBlIm40KzZpxYft2hN5ARkoqNik4RHtasoo67KMpG90KUAFo+eijXNy9m6DwcP589lmP5dMZHeYNIdAbjVjT06nYvDmd3nqLhvfdp2mlEC0pkTZfhcIZF/fs8aqfDTAEBdNq2DBtBSoG6AwG7pkwgXsmTMi6lhYfz5nNmwksW5aqbW8lPd1K9NYNnNlQmmtna3Fg7lwsTlbLmczt399rue6eMIHanTsT3qgRUkqunTlDaJUqhea14C0lOchCociT5JgYr6OgLLpSjNz6d5EsTuoPAsuWpWHPnlnnBqOeet26Ua9bNwDaPPUU0zp0KLB8CYbAQG55/PEcNf/CGzUqkLm0Rtl8FTccW774wuu+1ZrWo1rzkl3IVEsqt2yJsVQpn3Nf54UhKIj+s2a5LLZaVLF7O/hns09FISj8QrSXJgdDUBAtHhyssTQlG73JxN1ffIHOoMFaS6dDZzRSsUULun70EWNOnXIra11RRausZpkIIfRCiD1CiKWu2qqVr8IvVG/fnhOrVnnUx1iqFOXq1KHt6NEFJFXJpXaXLtjyyKnhLgFlytB72jRu6ttXQ6mKBhrbfJ8DDgEujd1q5avwC22efBJDcLBbbUtVqkSTBx7g3okTGbFjR7F9xPUna996y2U+lLyo37MnY86cYVx8fIlUvJk2X3cOVwghqgM9gSnuzK1Wvgq/EFK5MrePG8e6t992qRSklPSbOVObx+YblNMb3Au0yI4wGBi0cCF6o3sRcsURD/18w4UQO7OdT3YkBstkAvAykDNCJh/UylfhN85u3uzWaiwlJoad331XCBKVPE6cSOHhrjP44cwQttOTdNyv7vHYpk0lWvHCf+HF7hw4qhdnO7IUrxAiEoiRUu5yd26lfBV+w5Pk+LsmT3bdSJGDNWsu07TxGhatsXGSlqzicSbxf6Q4WZjpDAZq3H47z548SfV27fJtV1LQ0OxwG9BLCHEKmAV0EUL86qyDeo5T+I2IJ5/kyKJFbmU3Kwg3qZKMlJIxA+cwzPwJYY4srmdpwiKeYyMDuJupAOiDggipWJGIJ5+kw4sv3pCmHS3Ci6WUrwKvAgghOgNjpZROY99vvL+0oshQu1Mnqt16q1vJYtzdnFPYObTzDJFxLxLAf0mLarKfobzOz7xPD92PDFq0iIaRkX6U0v+oCDfFDcnVkyc5vXGjW23zSp6vyJ+tbz2PjpyuZXpsBHONuuxBHxBAfR+SGpUUCkL5SinXAetctVPKV+E3Fo8YYS/D6waVi1BxzKKOzWLh/OqlGMkdvi2w0px1tL9BTQx5ocKLFTcc7pgbMun81lsFJ0gJY9f332Mzm/OsAKLDxq23BNLlvff8IVqRw4aueIYXCyEeEEIcEELYhBAR1917VQgRJYQ4IoQo+BrQimJHXlWM8yMjMbEAJSk5xB09yjJHHt7spedxvA4MMjJyx3Z/iFZk0TK82BN8XfnuB/oC32e/KIRogr1eW1OgKvCXEKKhlNL7+EZFiUNvMrmdz1cVQ3WNNSOD2X365PCdzlFhWKej36xZ6m+ZDX9uuPn0X5BSHpJSHsnjVm9glpQyXUp5EogC2voyl6Lk0WywewlydCYTFZo0KWBpijcply/zdf36TqsMN+7Th8bFOAlOQSBBs/BiTymor8BqwNls5+cc13IhhBgphNgphNgZGxtbQOIoiiI93Yxa6/XDD2q15oIlI0eScPas0za3vfxyIUlTnPBf6XiX72ghxF9CiP15HL21EEBKOTkzXK9ChQpaDKkoJhgDA2k5dKjTNs0feoiWjzxSSBIVX45cV/L9ekKqVKF6W/XweT1ap5T0BJfqXErZzYtxzwM1sp1Xd1xTKHLQc9Ik0q5dy1N5lG/YkJ6TJvlBquKFtNmQNlu+94XBwKMeeJbcSEgE6SWsdPxiYKYQYjz2DbcGgNpiVeTCGBTEoIULSTh3jmPLlxP1559IKWk+aBA39e2rfFHdwOJi0/Leb74hrGHDQpKmeFFsqxcLIfoA3wAVgD+EEHullHdLKQ8IIeYABwEL8LTydFA4o3T16rQePpzWw4f7W5RihZSS31xsorky7dzoFMvwYinlQmBhPvc+AD7wZXyFQuGcU+vWOQ1WCQoPxxgUVHgCFTNUbgeFQuEV+375xWkV6Fsef7wQpSl+SARWm1K+CoXCQ6SLZPQdXnqpkCQpnkibID2tGIYXKxQK/1KjQwen9y/s2FFIkhRPpBRYLXq3DmcIIQKFENuFEP84Ui6842putfJVKIoxSZcuOb2fcuVKIUlSTJG4VKxukg50kVImCSGMwCYhxJ9Syq35dVDKV6EozrgwO1S46aZCEqR4IqXAYvZd+Uq7/SfJcWp0HE7/OUr5KhTFGKOTCh86k4kqKg+yCwQ2qzbVi4UQemAXUB+YKKXc5mwwpXwVimJMqhOzQtWIiHzvKRxIwH2zQ5yUMt8/qiOW4WYhRFlgoRCimZRyf37tlfJVKIoxZzZvzv+muD6VuiIXNgFp2qpBKWW8EGIt0AN72t08Ud4OCkUxJfXqVc5u2pTv/QvbVUS/W1jcPJwghKjgWPEihAgCugOHnfVRK1+Fopgy79VvsFqt+a6gbGYz6QkJBJQuXahyFSvsCX21oAow3WH31QFzpJRLnXVQylehKIZMn7iXP75fSVMX7VIuX1bK1xkaKV8p5T7Ao91NZXZQKIoR0mZjx/eTeWb0fi5SHzPGfNsKg4EyNWrke1+BXfma3Tw0RilfhaIYsWjYMH4f8wYplGYPd2PFSH6ZfPUmE2nx8YUqX7FDYg+PcOfQGKV8FYpiwLE//+TbRo345+efMabFMo5+jGY4RtIBXd7e/DYbO1Qyeudkmh183HDzBmXzVSiKKFJKTm/YwJrXX+dsNpcyARixYHShESxpaU69IRRoueHmMUr5KhRFECklix9/nP2zZ2NJSfFuECEIV+HFzvGj8vXJ7CCE+EwIcVgIsU8IsTDTz81x71UhRJQQ4ogQ4m7fRVUobhxOr1/PgTlzvFe8AFLS5umntROqJOJHs4OvNt9VQDMpZQvgKPAqgBCiCTAIaIo9yuP/HP5vCoXCDQ7Nn4/ZF8XrYMnIkVyJitJAohJMcVS+UsqVUspMsbZir1IM0BuYJaVMl1KeBKIAVbdaoXATfWAgQuf7fvjpdev4pmFDJjZtyqzevTk4f77TSsc3HDYgzc1DY7T0dngM+NPxuhpwNtu9c45ruRBCjBRC7BRC7IyNjdVQHIWi+NJyyBD0Jo1KmktJ3MGDHFm8mN+HDmX+Qw+5rIBxw1CUzQ5CiL+EEPvzOHpna/O6Q7wZngogpZwspYyQUkZUqFDB0+4KRYmkUosWdHn/ffSBgRhLlUIfoE2pG3NyMkcXL+bc1nxzfN9YFGVXMyllN2f3hRCPApFAV/nf1+l5IHtoTXXHNYVC4SbtX3iBZoMGcezPPzEEBlK7c2fObNrEmY0biVq+nKvHj3s1rjklheMrVlCjfXuNJS6GFFdXMyFED+BloJOUMvvuwGJgphBiPFAVaACoFEsKhYeEVq2aowJxs4EDaTZwIACJFy6w8qWX2D9rFnhox7129iwZycmYSpXSVN5iSXF0NQO+BUKBVUKIvUKI7wCklAeAOcBBYDnwtCPRsEKh0IjQqlXpN2MGb1utdBg3zqO+e6dN45Py5fl7/PgCkq6YoJHZQQhRQwixVghx0FFA8zlXU/u08pVS1ndy7wPgA1/GVygU7tH9o4+o1707v3Tt6nYfW0YGK198kYSLF7n7s88KULoijA1I1WQkC/CilHK3ECIU2CWEWCWlPJhfB5XbQaEoIdTt0oUu77/vcb+tn3/Ouv/9T3uBigMSsLp5OBtGyotSyt2O14nAIfLx8MpEKV+FogTR8fXXGbZ5M6WqVPGo3/p33iHu6NECkqqI477ZITzTLdZxjMxrOCFEbey5fZ0W0FTKV6EoYdTs0IGxFy4waMkShAe+wn889VQBSlVE8czmG5fpFus4Jl8/nBAiBJgPjJFSJjibWilfhaKE0igykvunTnW7ffTu3QUoTRFFQz9fIYQRu+KdIaVc4Kq9ymqmUJRg6vfo4X7jG7HacWZ4sY8IIQQwFTgkpXTLhUStfBWKEkxweDjlGzVyq2252rULVpiiijYr39uAIUAXh9vtXiHEvc46qJWvQlHCGf7333xavrzLdjdk+kntCmhuwp7n3m3UylehKOEElSvHwN9/d9qmbJ06tBgypJAkKkKoApoKhaIgady7N+MSEqjWrl0u227d7t0ZffgwemP+lZBLLBr5+XqDMjsoFDcIAaGhDN+6leSYGM7v2IHQ6ajRvj2BZcu67lxSKa6JdRQKRfGjVMWKNOzZ099iFA0kWoUXe4xSvkWQaBL5mi2c4AoGdHShHkO5Bb2yEikU2pJpdvADSvkWMeJI5lmWYMOeGtmCjWUc4ShxfIwHPpsKhcI1yuygAHu58G+i5mKrJcCU818TxWVOE08tbmD7nB/ISEpi/6xZxB48SOVWrWj6wAMYAgP9LZZCK5TyVVjS0/n17rs5+/wt0KB1nm22c0Yp30Ig6dIlzv39N3GHD7P+3Xexmc3YLBZMISGsfeMNhm/bRkjlyv4WU6EFma5mfkAp3yJA3JEjzO7bl7iDBzHdUxl6tICA3G4/VQj1g3TaELN/P2e3bCGkShXq9+hRJN2apJSsffNNtnzxBTaLBWnJuSTKSErCkpbGihdfpN8Mj8sVKooqxdHmK4R4D3uZeBsQAzwqpbzgiHP+CrgXSHFcvwGzduRPyuXLSCnZ8tlnbPn8c1KbVSflia6Yoi4hzFZkduUrJUah5zZq+01eb0iJi2P9e++xY+JEpNX+DhcGA0Hly/PYxo2ENWzoZwlzcnTJEv4ePx5rWv7B/jaLhSOLFhWiVIoCRaPcDt7g68r3MynlmwBCiGeBt4AngHuw121rALQDJjl+lnjijh5l748/cmn/fnuBQ5uN0GrVSE9IIPnSJcrUrElybCzxp05hM5uRUnLhm0e4OqwTCBAWK7p0MyI+BWv5UqDXEZBq45YBU/lg/aPoTCZaPfYYXd5/n4DQorsSTrxwgYlNm5IeH5/jurRYSImJYXbfvjy1f79m853bupXt334LQLV27UiJi0NvMtFs4EDK18+34EoWZzZtYv5DD2FJde13pNPrfZZXUUQormaH6/JVlgIyqxf3Bn52VDPeKoQoK4SoIqW86Mt8RZnTGzfyW69euZQNwOUjR7JeJ5w9m/OmgErvLuTq452RQSb7H9Bqw3T+HPU6vENwuXJY9kWR6Ghuzchg+9dfc3jBAkYfPYoxKKjAfidfWP3663n+LTK5fOwYV0+coFzduj7NI6Xkt169OLZ0ada1fx0mAWEwsPGDD7h7/Hginngi3zH2/vwzS0eNcrrizUKno9ngwT7JrChC+NHVzGfHUSHEB0KIs8BD2Fe+YC+fkV3LnCOfkhpCiJGZmeFjY2N9Fccv7P3xR3664w6nyiY/hASRmkGZBTv+u6jXYa5fGXQCy76oPPslnDuXpWSKIocXLnR632a1YnFH2blg+fPP51C82ZEWC5bUVFY8/zxJ0dF5tslITmbJ8OHuKV6gfL16dP/0U6/lVRRBNMrn6ykula8Q4i8hxP48jt4AUsrXpZQ1gBnAaE8FkFJOzswMX6FCBc9/Az9jTk1lUbbS3t6gS07HeDoOncXK7Su38PozH/PmE+/RMjDNaZqkI4sX+zRvQSJdlDI3BAQQ3rixT3PsnjKF7V995bKd0Os5mo+CXvDww9jM7j13lm/QgNFHjhBQurRHciqKMNomU58mhIgRQrhlT3NpdpBSdnNnIOzKdxnwNnAeqJHtXnXHtRLHnqlTQUrXDV1ga1aNLx4aR/UzF7MUbvOycPVmmPwvpOfxaFSmZk2f5y0oqrVrx8m//sr3fu/p0xE67x+8bFYryzxIgSjysNMenD+fIy6yfWUSWq0aT/77L+JGTDhektF2w+0n4FvgZ3ca+2R2EEI0yHbaGzjseL0YeETYuRW4VlLtvdfOnPF9ECG4d9sOqmVTvI7LlAuAztXz7lOU86/eO3Fivso1uGJFmvXv79P4V0+cwJqR4VZbabXSqFcvAGIPHmR2v358VrEicz2Q4ZmjRzEEBHglq6IIo+HKV0q5Abji7tS+ejt8LIRohP374zR2Twewr4DvBaKwu5oN83GeIkul5s19HiNASLqt35bnN6EQ0CIcVpzOef2+77+nwk03+Tx3QRHesCFDVq3it969MSclZV0Pa9SIETt3+jy+s0d/CSRSDr0xgNK6q/SeNo3gsDBiDx5kSrt2ZCQne/S0ct+UKRiDg32WWVFEcd+eGy6EyP7mnZxXEU138dXboV8+1yVQdJdlGtJk4EAWPvoouLBxOsNmgyCLJd88+JlqQuh0tH/pJe58910MHlSl9Rd1unTh1YQE4k+dIiUujvL16hHkRkUFdwipVImQqlVJunAhx3XpOEpzlQxzILJJR5oOHAjYPTA8UbyGoCAe3bCBahERmsisKIJ45moWJ6XU7M2g0mT5iMFk4l6Hf6k3GHXQuQboRN46wSZhr8MJROj13PHaazkUb+yhQyweMYJFjz9OzMGDXstRUAghKFenDtXatNFM8WYyYscOhMG+frBh/xwJ/ntTm0gj4+A6Zj44gkktW9rtux6seJ/Yu1cp3pKOSqZevGnz5JNkJCby1yuvuNXeEByMEAK90UizsmbaVEgm3Qp6AddvC8WlwvpzdsVbq2PHHI/bi4cPt2/4Odg7bRqmsmUZtm4dlVu21OJXK3JIm40zmzeTEhtL9fbtGRcfz4KHIpvU5wAADRtJREFUH+avRceoJKMwkZ6jvQELR2f/iA7PNkVrdOxY5CLwFAWAHxPrCKnBTr1WREREyJ0a2AP9hTklhe3ffsu/M2cSd/Qo0mIhoEwZGt9/Px1fe41yderk2S/1wjms509S6qaWxCz8lWv/9z/SLsfxb6zkRIoRQ0AgweHhDNu4kdLV7O7S57ZuZWr79vnK0roidKtvwvbY2wQ//lqB/L6Fic0m+WTcOmI/H0ygTAAERp2FCveNYPisCXw45DvM814mUKPM2K8mJ2NSdt4ijRBil69mACEiJAY3dY7F+XxCiN+AzkA4cAl4W0o5Nd/2SvkWXc7v2MHF3bspW7s2dbt1yxHW+lvv3hy9zs832AAdqkKDsvbXJj0YBNha3o5h8sZClT322En++fUX9k37gcTzdi/D4LAwqrZty8nT6Sw/24ZoXWMa1LDQq/EBalaycOp0OjH7/kGXHEuGzcC1sjdz05CRDHu5M++/d4zkj++kHNE5TONmjOwMHspNw54gaGJHAjRQvsJgYNDChTSMjPR5LEXBoZnyFW7qHOn7fNlRZociTLU2bajWpk2e96zpOR+vgw3wZAsINIAhmyVfStDv2wSblsLtBa9M9v+1nTH3TaFd2k8YHTsZmcoyJS6OY8uW8Scj2UkHbBg4EW9h9b/lGMZLVOFEVlsjEH7lKNHvLOCRrwdz+loYPa9TvPZ2ZtqlTOHS/61nI49zF1MxkuaxmSHHmIGBWN0MvFCUAPy0/lTKt5jS5umnOb5iRdZ5h6q5FS/YXdWkBN5/DJbH+Dxv9LHTzH3/N6IPHCY52cJpwy1UbnkzAx6uS8qyCaz4Zg63cREdkEowFgwcoBMZBNOAHVThBD34gf10IoVySAyYMZBCmZw+zo6fBiw0ujqLRuTrDIIOqCyPcS8n2MG9hBBPY7Zg8HKXxJKeTt2uXb3qq1C4i/J2KKY0uu8+atx2W9Z5g7K5FW8mQgAJl32a7+jGXbxQvjVfNmzOpZ/f4tKubUw63Ift+3VcnPEhM+7pwU/f7KScQ/FKIIBUTJhpxgb+oTPT+JxlPAFIIsnpIVKF4/nObVfPzleiAtBj5RZWsoN7uUw1MvDOHc8QFKRCiBUFjlK+xZjHNm2ihyO3QaKrYC+j937Bpw9dYHqnjqRdjSOJMiRRjtm8QQ0O8SjjaMFaanCERmxDOt5SdpcvSQCpBJPAA3yKmUD20J0zNKM+OdM7p7pMFO9eWK+RdDoymzJcRo/FqyfKjMREbBY/bYErbhiU2aGY8//t3XtwVPUVwPHv2U2WECnyCI+MgAmvlvCooEWKhZbQqcHaQGeAiTIFlRHaImKltgIOfxRH2ilTK9OWlj6ojo6glQqtoKXi1BkQIRIkgMXyqEgIApU3JQmb0z/uTboJeWyyN7n7OJ+ZHfY+cvf3mwwnv7339zvn9ocfZvT8+Wyb9AX6fvoeoQZSzSog32h98p9VRY9xQgfzKgu5SieqCVBNkEJW1pnald5IsAugdKeMzpzmAt3ZywT6sb/2eBoVlNOfrEbSf4QJ0aFTJtWXms8aJ8BA9rS0i3V8JjubQJr910gN1fhVO95GvklARPjS68WcyxtPVbVzj7f2BciA4fC95rN/NeZC6TusZSkXyaKKDMKEuIFzZHL++rY0cg1FCLoTKgXlBIPowCXSqGQwO/kc22tXp9W8qnFmM4RH38+sN14j0A65FYKhEPnLl7f555h4UbPELZqXt+zPexLp+cd/UHHsCEdWLqPjmaNkDRpMaMpsGBpbEZEDOpbqess/KumIEP2S6kt05Sy9SaeCEWylC+Xcw4/oSjldgufoMWQIgdzb+I/0o/vNvUg/e5QKDTFu9lSGTRgJwGMnT/LilCkce/ttTzLJ1SdpaRSuWcOIe+/1/NomXvm3ysKCb5Lp0K8/A1as8fSaF0N9CVfWvWdcSSYfMprB7KrzMOwaaREjXKgkhBJgPd8nSBU5gQPcULiYLw87wY2Z17h5/Hj6jh0bVarGDHf13tGtW9m1ahVXz53j1L59XG4kUXpLBEMh5paU0CMvL+ZrmUTiXx0hC76mWYVFn6X0uf9SRd2SRRuZzyyWkMVxwqQRpIrDwTHcueg7lBXv5tK2V6gOdqSk+2z6Z9/O8rk5zJhRGHNO3Nz8fHLz8wH4y4MPUrJmTW2BztbokpPDN59/3gJvSvIv+NoKN9Osq1fDDOn3KsdOp1NN/ZLvSjfKyOIET/66gGlzo7vFca2igmB6ekwJ1QE+PXyY39xyC5URaSujJcEgE5YtY9yiRTG1wfjDmxVuwxXWR3n2YE9XuNkDN9OsjIwgew9PZsmSgXS/sSZ/mPPK6lzF8qXZHKhaElXg/Xj7dlaNGMFTmZk81akTr82bF1Mtt24DBvDA9u3kTpwYVSAPhEIQCNBz+HDmFBdb4E15HmZTbyEb+ZpWqawMk5YmBFowcj1z8CCrR42i6sqV2n1pGRkMKCigqJmCm9EqLymhbOdOqi5fZt+6dZzcswdE6Dd2LOOeeAINh+mRl1eboMgkLm9GvkMVXozy7M/HX24HEVkIrAB6qOoZcW7qPYNTzeIKcJ+q7m7qGiaxhBqaUNyMd1asoKreKPfa1asc2rSJ88eOeVKTLnvkSLJHOrMjvvjoozFfzyQ7/2Y7eFE6vi/wNSCymNkkYJD7mgOsivVzTOL7pLS0wYof4cpKPtq2zYcWGePdPF8RKRCRgyJySEQeb+58L+75Pg38gLq5gSYDz6ljB9BFRLI9+CyTwLoNHNjosZZUIjbGO97c8xWRIPBLnIFnHnCPiDQ5fSbW6sWTgTJVfb/eoZuAjyO2j7v7GrrGHBEpFpHi06dPx9IcE+d633pro8cqzp5tx5YYU6NmeXE0ryaNBg6p6hFVrQTW4gxCG9XsPV8R+TvQu4FDS4DFOLccWs2t/rkanAdusVzLxLfMrKwmj2t1dcxTz4xpmRbN822qenFDA84mp/80G3xV9asN7ReR4UAu8L47ab4PsFtERgNlQN+I0/u4+0wKy5s6lQ0zZzZ+QoyLL4xpnagfuHlavbjVsx1UtRToWbMtIv8GbnNnO2wEHhKRtTjR/7yqlsfaWJPYQh07IsFgg6vRaoqKGtO+PFvh1uIBZ1t9x9sEHAEOAb8FvttGn2MSzCNlZdePcAMBFhw96k+DTIrzbLbDLmCQiOSKSAgoAjY29QOe5XZQ1ZyI9wrY42tznc69erE0HGb/yy9zePNmBkyaxNBp02zUa3zizTxfVb0mIg8BbwBB4A+qur+pn7HEOqbdiQjDpk9n2PTpfjfFpDzvkqmr6iacb/1RseBrjElhllLSGGN8YMnUjTHGBzbyNcYYH9jI1xhjfOBf9eK4yucrIqeBjyJ2ZQFnfGqOX1Kxz2D9TjVe9PtmVe0RywVE5HW3LdE4o6oFsXxenc+Op+Bbn4gUe7mcLxGkYp/B+u13O9pbqvY7kmUxMcYYH1jwNcYYH8R78F3d/ClJJxX7DNbvVJOq/a4V1/d8jTEmWcX7yNcYY5KSBV9jjPFB3AZfEVkoIioiWe62iMhKtzLoXhEZ5XcbvSQiPxWRf7p9+7OIdIk4tsjt90ERudPPdraFllZ9TVQi0ldE3hKRAyKyX0QWuPu7icgWEfmX+29Xv9vqNREJikiJiPzV3c4VkXfd3/k6NwduSonL4Jui5ei3AMNUdQTwIbAIwK2AWgQMBQqAX7mVUpNCa6q+JrBrwEJVzQPGAPPcvj4OvKmqg4A33e1kswD4IGL7J8DTqjoQOAvM9qVVPorL4EsKlqNX1b+pas0i8x04ZUjA6fdaVa1Q1aM41UFG+9HGNtLiqq+JSlXLVXW3+/4iTjC6Cae/z7qnPQtM8aeFbUNE+gBfB37nbguQD/zJPSXp+hyNuAu+XpSjTwIPAJvd98ne72TvX4NEJAcYCbwL9IqocXgS6OVTs9rKz3EGU9XudnfgXMRgIyV+5/X5klinrcvRx6um+q2qG9xzluB8PX2hPdtm2o+IdAJeAR5R1QuRJZRUVUUkaeZ/isjdwClVfU9EvuJ3e+KJL8E3VcvRN9bvGiJyH3A3MFH/PwE74fvdjGTvXx0iko4TeF9Q1fXu7k9EJFtVy91baaf8a6Hn7gAKReQuIAPoDDyDc9swzR39JvXvvDFxddtBVUtVtaeq5rgFOY8Do1T1JE4l0JnurIcxJFk5ehEpwPlqVqiqVyIObQSKRKSDiOTiPHDc6Ucb20iLq74mKvde5++BD1T1ZxGHNgKz3PezgA3t3ba2oqqLVLWP+/+5CNiqqjOAt4Cp7mlJ1edoJVI+303AXTgPnK4A9/vbHM/9AugAbHFH/TtU9duqul9EXgIO4NyOmKeqYR/b6anWVH1NYHcA3wJKRWSPu28x8GPgJRGZjZNSNRUqi/4QWCsiTwIlOH+UUootLzbGGB/E1W0HY4xJFRZ8jTHGBxZ8jTHGBxZ8jTHGBxZ8jTHGBxZ8jTHGBxZ8jTHGB/8DL1d/WbJ6sqEAAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [74 33 79 79  0 79 74 35  0 74  0 35 74 30 33 35  0 33 33 79 79 17 74 74\n",
            " 30 33 74  0 79 79 74 33 54 74 79  0 79 79 13 79 74 79  0  0 17 33 33 11\n",
            " 33 74 74 79  0 79 92 33 35 33 20 35 74 85 74 33 33 13 35 92 18  4 79 74\n",
            " 54  0 79  8 33 33  8 74 74 88 35 79 33 33 35 74 33 54 30 35 13  8 32 60\n",
            " 74 79 74 85 33 13 79 54 74 79 35 33 79 74 79 33 54 13 35 33  8 35 54 79\n",
            " 18 11 33 13 74 74 54  8 13  0  0 33  0  0 79 79 79 33 74 74  0 79 33 74\n",
            " 79  0 54 33 79  0 79 79  0 91 60 11  0 30 85 64 79 33 35 33 33 79 79  0\n",
            " 30  0 13 11 33 35 79 33 74 30 64 79 35 33 79  0 33 60 85 74 33 74  0 35\n",
            " 54 30 79 33 33 13 35 60 33 60 33 33 33 33  8 33 35 33 79  0  0  0 30 35\n",
            " 79 79 74 74 35  0 13 79 35 74 54  0 35 79 54 79 74 74 33 79 74 60  0 85\n",
            " 33 85 74  0 74 60 33 13 18 33 79 33 54 79 74 85 33 88 33 30 33  0 33  0\n",
            " 60 74 33 74 33 54 33  0 74 79 13 54 79 33 79 60 33 74 74 74 35  0 33 33\n",
            " 91 85 79 74 74 13  0 33 35  0 35  0 30  0 33 74 33 35  0 79 35  0 33 33\n",
            " 35 79 35 35 33  0 74 35 35 35 85 33 60  0 60  0 79 17 35 35 79 79  0 13\n",
            " 54 33 33 74 74 74 54 74 35 17 54 74 13 74 35 30 30  0 33 13 33 18 74 35\n",
            " 79 92 54 33 79 33 33  0 79 35 33 60 33 60  0 54 13 35 79 33 32 60 91 13\n",
            " 11 64  0 30 11 91 33  0 35 74 74 33 35 33 33 33 91 30 85 33  0 54 64 91\n",
            " 33 33 33 33  0 91 33 35 79 79 33 33 79  8  0 92 33 33  0  0 74 13 74 74\n",
            " 35  8 30 60 33 33 54 60 60 79 79 79 60  0 33 33  0 54 33 33 35 33 64 33\n",
            " 79 33 79 35 33 79  0 30 79 35 33  8 33 13 30 79 33 74 33 30 33 33 13 60\n",
            " 60 33 33 33 30 33 54 74 85 74 30 13 54 33 30 54 79 33 35 11 35 92 92 79\n",
            "  0 35 13 60  0 79 13 33 11 35 60 74 79 35 79 79 35 74 79 30 18 35 74 13\n",
            "  0 35 60 33 74  0  0 54 33 74  0 74 74 35 35 79  0 35 74 79  0 35 30 35\n",
            "  0 33 33 35 79  0 79 74 33 60 79  0 13 30 64 33 74 74 74 79  0 33 79 79\n",
            "  0 35 13 35 33 79 35 13 33 30 74 54 33 33 74 33 54 79 74 13 74 33  0 79\n",
            " 30 33  0 11 54 79 35 35 33 35 30 74 79 35 35 79 74 79 79 79 79 54 91 79\n",
            " 85 33 18 74 79 13 33 35 20 11 54 33 74 33 33  0  0 33 91 13  0 54 35 79\n",
            " 79  0 74 13 79 35 79 35  8  0 35 60  0 74 33 30 74 30 30 33 33 64 35 11\n",
            " 35 30 17 30 79 92 79 35 60 79 79 13 74 79 74 74 35 30 74 33 35 74 13 74\n",
            " 33 64 33 79 13 33 13 33 35 33 54 79 13 54 60  0 74 74  0 35 79 29 33 33\n",
            " 33 30 35 35 92 33  0 33 74 30 33 33 33 74 74 33 13 33  0 33 35 35 64  0\n",
            "  0 33 29 79 35 79  0 74 33  0  0 64 79 35 33 85 33 33 85  0 74 54 74  0\n",
            " 33 33  0 79 74 64 30 74 33 88  0 17 74  0 35 79 54 79 29 79 33 33 33 33\n",
            " 74 54  0 33  0 33 60 33 33 59 35 20 74 33 33 74 30 74 33 79 74  8 13 79\n",
            " 74 79 79 79 30 33 35 13 30 79 13 30 79 79 18 33 13 54 33 79 35 13 35 79\n",
            " 79 79 33 33 79 79 74 33 35 85 33 54 79 79  0 33 54 35 30 60 74 30  0 74\n",
            " 33 13 33 74 20 79 13 30 30 33 33  0 11 79 79 33 79 35 79 79 33 30 85 33\n",
            " 60 33 33 33 33 74 74 60 33 33  0 13 33 74 74 35 85 74  0 74 85 79 79  0\n",
            " 79 74 74 30 60 30 30 33 74 35 74 79 74 85 79 79 30 33 74 35 74 35 74 74\n",
            " 17 33 30 60 35 74 33  0 33 30 13 74 30 79 79 54 35 33 35 79 35  8 35  4\n",
            "  0 85 13 74  0 54 33 35 74 13 79 74 35 74 33 35 33 54 74 54 30 33 30 79\n",
            " 74 79  0 13 33 85 74 33 79 33 13  0 33 30 33]\n",
            "iter  800  loss  0.41372889280319214\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [74 35  0  0  0  0 74 35  0 35 74 74 74 30 33 35  0 33 33  0  0 35 74 74\n",
            " 30 79 35  0 79 79 74 33 30 74 79  0  0  0 13 79 74  0  0  0 35 79 79 17\n",
            " 79 74 74 79  0  0 92 33 74 33 17 35 74 35 35 79 33  8 35 18 18  8 79 74\n",
            " 33 74 79 26 33 33  8 74 74 17 35 79 33 79 74 35 33 54 30 74  4  8 17 79\n",
            " 74 79 74 85 33 13 79 54 74 79 85 33 74 35 79 33 54 13 35 33 13 35 54 79\n",
            " 18 17 79  8 74 74 54  8 13  0  0 79  0  0 79 79  0 60 74 74  0  0 35 30\n",
            " 79 74 54 79 79  0 74 79  0 17 60 17 74 30 91 17 79 60 35 33 60  0 79  0\n",
            " 30  0 13 17 33 74 79 33 74 13 17  0 85 79 79  0 33 79 35 74 33 35  0 35\n",
            " 13 30 79 33 33 13 74 60 79 60 60 33 33 33  8 79 35 79 79  0  0  0 79 35\n",
            " 79 74 74 74 35 74 13 79 85 74 33 79 35 79 54  0 74 74 33 79 74  0 79 35\n",
            " 60 85 74  0 74 79 33 13 18 79 79 33 54  0 35 91 60 17 33 33 79  0 79 74\n",
            " 60 74 60 74 33 33 74  0 74  0 13 33 74 33  0 60 79 74 74 74 35  0 33 79\n",
            " 17 91  0 74 74 13  0 79 35  0 35  0 30  0 79 74 79 85  0 79 35  0 54 33\n",
            " 35  0 35 35 79  0 74 74 85 35 85 33 79  0 60  0  0 35 85 35  0  0  0 30\n",
            " 13 33 33 74 74 74 54 35 85 74 33 74 13 74 35 33 60  0 33 13 33 18 35 35\n",
            "  0 18 33 60 79 33 33  0  0 35 33 60 33 60  0 54 30 35  0 33 17 60 17 13\n",
            " 35 17  0 33 17 35 33  0 85 35 74 79 35 60 79 33 35 33 85 33  0 33 17 17\n",
            " 74 79 33 60  0 17 33 85  0  0 33 33 79  8 74 92 33 33 74  0 74  8 74 74\n",
            " 35  8 30 60 33 33 54 60 60 79 79  0 60  0 33 74  0 33 79 79 35 79 17 74\n",
            " 79 74 79 74 33 79  0 30 79 35 33  8 60 13 30 79 79 74 33 60 33 33  8 79\n",
            " 60 33 33 33 33 33 54 74 35 35 30 13 92 79 60 33  0 79 74 35 35 18 18  0\n",
            " 74 35 13  0 74 79 13 79 17 74 79 74 79 35 79  0 85 35  0 13 18 35 35 13\n",
            "  0 35 60 33 35  0  0 54 33 74  0 74 74 35 35  0  0 85 74  0  0 35 60 74\n",
            "  0 60  0 35 79  0  0  0 79  0 79  0  8 30 17 79 74 74 74 79  0  0  0 74\n",
            "  0 35 13 35 33 79 74 13 60 60 74 33 33 79 74 33 13 79 74 13 74 33 74  0\n",
            " 30 33  0 17 33 79 35 35 33 35 33 74  0 35 35  0 35 79 79 79  0 54 74  0\n",
            " 35 33 18 74  0 13 60 35 17 17 54 33 74 33 79  0  0 60 17  8 74 13 35 79\n",
            " 74  0 74  8 74 35 74 85 26 74 85 79  0 74 79 60 74 30  4 79 35 17 85 17\n",
            " 35 30 74 30  0 54 79 35 79 79  0  4 35 79 74 74 35 30 74 79 35 74 13 74\n",
            " 33 17 79  0  8 33 13 79 35 79 92  0  8 54 79  0 74 74 74 35 79 35 33 79\n",
            " 33 30 35 85 92 35  0 74 74 30 33 33 33 74 74 60  8 79  0 60 35 35 17  0\n",
            "  0 33 17 79 35 79  0 35 33  0  0 17  0 35 79 85 79 33 45  0 35 54 35  0\n",
            " 79 60  0 79 74 17 33 74 79 17  0 74 74  0 35  0 54 79 74 79 79 79 33 33\n",
            " 74 33  0 79  0 79 30 33 54 74 74 17 35 79 79 74 30  0 74  0 74  8 13 74\n",
            " 35 79 79  0 30 33 35 13 30 79  8 33 79  0 82 33  8 33 33 79 35 13 85 79\n",
            " 79  0 79 79 79  0  0 33 35 85 79 54 79 79  0 79 54 85 30 60 74 60 74 74\n",
            " 33 13 33  0 17  0 13 13 30 33 33  0 17 79 79 33 74 35 79  0 79  4 35 33\n",
            "  0 33 79 74 33 35 35 79 33 33  0  8 33 74 74 35 35 74  0 35 85 79  0  0\n",
            " 79 74 74 30 60 60 30 33 74 74 74  0  0 35 79 33 60 79 35 45 74 35 35 74\n",
            " 35 79 30 60 85 74 33  0 33 30  8 74 30  0 79 54 35 33 35 79 35  8 74  4\n",
            "  0 35 30 35 74 54  0 35 35  8 74 35 35 74 79 85 33 54 35 54 60 33  4  0\n",
            " 74 79 74 13 33 35 35 33 79 33 13  0 33 30 33]\n",
            "iter  900  loss  0.7672337293624878\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n",
            "  FutureWarning,\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "prediction  [17 35  0  0  0 74 35 35 74 35 35 35 35 30 33 35  0 79 33  0  0 45 35 74\n",
            " 30 74 35  0  0 74 35 79 33 74  0 35  0  0 30 79 85  0  0  0 45  0 79 45\n",
            "  0 74 35  0 74  0 92 79 35 79 32 85 74 85 17 79 33 13 91 18 18  4 79  0\n",
            " 33 35 79  8 33 33  8 74 35 86 45 79 79 79 85 85 79 54 30 85 30 13 86 79\n",
            " 35 33 35 45 33 13  0 33 35 79 45 79 85 35 79 33 54 30 35 33 13 85 54 79\n",
            " 18 32 79  4 74 74 54  8 30  0  0 79  0  0 79  0  0 79 35 35  0  0 45 35\n",
            " 79 35 35  0 79  0 35 79  0 32 79 32 74 60 32 64 79 79 45 79 79  0 79  0\n",
            "  0  0 30 64 33 85 79 33 35 60 64  0 85 79  0 74 33 79 85 74 79 35  0 45\n",
            " 33 30  0 79 79 30 17 60 79 79 79 33 33 33 13  0 85 79 74  0  0 35 79 45\n",
            "  0 74 35 35 45 74 30 79 45 74 33  0 85  0 33  0 35 74 79  0 45  0  0 35\n",
            " 79 45 85 74 35  0 79 13 18 74  0 79 33  0 35 45  0 64 33 79 79 74  0 74\n",
            "  0 35 79 74 79 33 74  0 74 79 60 33 74 79 74 35 79 35 85  0 35  0 33 79\n",
            " 45 45 79 74 74 30  0  0 85  0 85 74 60  0 79 17  0 45  0 79 35  0 33 79\n",
            " 85 79 35 85  0  0 35 85 85 85 45 79 79  0 74  0  0 45 85 85 35  0  0 54\n",
            " 54 33 33 74 35 74 54 85 85 45 33 74 13 35 85 33 60 74 33 30 79 92 35 85\n",
            "  0 92 33 79  0 79 33  0 74 91 33 79 33 79  0 54 54 35  0 33 32 60 64 13\n",
            " 45 64  0 79 32 85 33 74 45 45 35  0 35  0 79 33 91 79 45 33 74 33 32 32\n",
            " 35  0 79  0  0 32 33 85  0  0 79 79 79 13 74 92 79 79 17 74 74 13 74 35\n",
            " 85 13 30 79 33 33 33 79 79 79 79  0 79  0 79 74  0 33  0 79 85 79 64 35\n",
            " 79 74  0 85 33 74 74 79  0 35 33  8 79 60 60 79 79 35 79 74 79 33  8  0\n",
            " 79 79 33 33 79 79 54 74 45 35  0 30 92 79  0 33 35  0 45 85 91 18 18  0\n",
            " 35 85 33  0 74  0 60  0 45 17  0 35 79 85 79  0 45 45 74 30 18 85 45 33\n",
            "  0 85 74 33 85  0 74 33 79 74  0 35 35 35 35  0  0 45 35 79  0 91 79 85\n",
            "  0 79  0 85 79  0  0 74 79 74 79  0 13 60 64  0 35 17 35 79 74  0  0 35\n",
            " 74 45 30 85 33 79 35 33 79 79 74 33 33 79 35 79 54 79 74 30 35 79 74  0\n",
            " 60 79  0 32 33 79 35 85 79 85 79 35  0 85 85 74 35 79 79  0  0 92 91  0\n",
            " 45  0 18 35 74 33 79 85 64 45 54 79 35 33 79  0 74 79 91 13 74 30 85 79\n",
            " 74  0 35 13 74 45 74 45  8 35 45  0 74 35  0  0 85 79 30 79 85 64 85 32\n",
            " 85 35 85 30  0 92  0 35 79  0  0 30 45  0 74 74 85 33 35 79 45 35 30 35\n",
            " 35 64  0  0 13 33  8 79 35 79 92  0  8 54  0  0 74 74 74 85 79 45 33 79\n",
            " 33 79 85 45 92 85  0 35 74 60 79 33 35 74 35 74 13 79 74  0 85 17 64 74\n",
            " 74 33 32 79 85 74 74 35 33  0  0 64 74 35 79 45 79 33 32  0 17 33 35  0\n",
            "  0 79  0 79 35 64 79 74 79 86 74 35 85 74 45  0 54 79 17 79 79 79 33 79\n",
            " 85 33  0  0  0 79 60 33 54 35 35 64 35 79 74 35 60 74 74  0 35  8 13 85\n",
            " 35 79 79  0 79 33 35 60 60 79  8 33 79 35 18 33 13 74 33 79 85 60 45 79\n",
            "  0  0  0 33 79  0  0 79 45 45 79 33 79 79  0  0 54 45  0  0 35 60 74 74\n",
            " 79 30 79 74 32  0 13 30 60 79 79  0 64  0 79 79 74 85 79  0 74 35 45 79\n",
            "  0 79  0 74 33 85 85  0 79 79  0 13 74 85 35 91 85 35 74 91 45  0  0 74\n",
            " 79 35 35 79  0 79 35 79 35 35 74 74 74 45 79 33  0 79 35 32 35 85 35 74\n",
            " 45  0 79 79 85 74 79  0 79 35 13 35 79  0 79 54 85 79 85 79 35  8 17  4\n",
            "  0 45 13 85 74 54  0 85 17 13 35 85 35 74 79 85 79 33 17 54  0 33 30  0\n",
            " 35 79 74  8 33 45 85 79 79 79 13 74 33 60 33]\n",
            "iter  1000  loss  0.413177490234375\n"
          ]
        },
        {
          "output_type": "error",
          "ename": "KeyboardInterrupt",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-77-be3ed66bbe14>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     20\u001b[0m     \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m\u001b[0mclustering_model\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_on_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mp\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m     \u001b[0mindex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mindex\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mindex\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;34m<=\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 22\u001b[0;31m     \u001b[0mgc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollect\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize = (15,9))\n",
        "n = 0\n",
        "for i in range(9):\n",
        "    n += 1\n",
        "    plt.subplot(3,3,i+1)\n",
        "    plt.subplots_adjust(hspace = 0.75, wspace = 0.6)\n",
        "    plt.imshow(x[n])\n",
        "    plt.title(\"prediction = \"+str(y_pred[i]))"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-08-01T20:39:18.848337Z",
          "iopub.status.idle": "2022-08-01T20:39:18.849066Z",
          "shell.execute_reply.started": "2022-08-01T20:39:18.848796Z",
          "shell.execute_reply": "2022-08-01T20:39:18.848822Z"
        },
        "trusted": true,
        "id": "lLwuBLRW6OFE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 553
        },
        "outputId": "eb9c4606-69df-40ef-d398-3fb8b1d0ee58"
      },
      "execution_count": 80,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x648 with 9 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}