VIt_model_2-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "48530309",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorflow-addons in c:\\users\\hp z book\\anaconda3\\lib\\site-packages (0.22.0)\n",
      "Requirement already satisfied: typeguard<3.0.0,>=2.7 in c:\\users\\hp z book\\anaconda3\\lib\\site-packages (from tensorflow-addons) (2.13.3)\n",
      "Requirement already satisfied: packaging in c:\\users\\hp z book\\anaconda3\\lib\\site-packages (from tensorflow-addons) (23.1)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install -U tensorflow-addons\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c1fc01e1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: imutils in c:\\users\\hp z book\\anaconda3\\lib\\site-packages (0.5.4)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install imutils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9e465df0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: opencv-python in c:\\users\\hp z book\\anaconda3\\lib\\site-packages (4.9.0.80)\n",
      "Requirement already satisfied: numpy>=1.21.2 in c:\\users\\hp z book\\anaconda3\\lib\\site-packages (from opencv-python) (1.24.3)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install opencv-python"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1ea0d8cb",
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'tensorflow'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[11], line 2\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mnumpy\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mnp\u001b[39;00m\n\u001b[1;32m----> 2\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mtf\u001b[39;00m\n\u001b[0;32m      4\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mnumpy\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mnp\u001b[39;00m \n\u001b[0;32m      5\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mpandas\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mpd\u001b[39;00m \n",
      "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'tensorflow'"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "\n",
    "import numpy as np \n",
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2409ecb5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "import tensorflow_addons as tfa\n",
    "import os\n",
    "from tensorflow.keras.preprocessing.image import load_img\n",
    "from tensorflow.keras.applications import MobileNetV2\n",
    "from tensorflow.keras.layers import AveragePooling2D\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.layers import Dropout\n",
    "from tensorflow.keras.layers import Flatten\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.layers import Input\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.applications.mobilenet_v2 import preprocess_input\n",
    "from tensorflow.keras.preprocessing.image import img_to_array\n",
    "from tensorflow.keras.preprocessing.image import load_img\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report\n",
    "import numpy as np \n",
    "import scipy.interpolate\n",
    "import re\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "\n",
    "from imutils import paths\n",
    "from PIL import ImageFile\n",
    "import pandas as pd \n",
    "ImageFile.LOAD_TRUNCATED_IMAGES = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bf44c0c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ['CUDA_DEVICE_ORDER'] = \"PCI_BUS_ID\"\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"0\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c4f7e6cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# initialize the initial learning rate, number of epochs to train for,\n",
    "# and batch size\n",
    "num_classes = 2\n",
    "input_shape = (224, 224, 3)\n",
    "\n",
    "df = r\"C:/Users/GB/Desktop/Taliqa/dataset/DS\"\n",
    "CATEGORIES =  [\"Normal\", \"STONE\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0a96b029",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'os' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[7], line 4\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# num_classes = 2\u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;66;03m# input_shape = (224, 224, 3)\u001b[39;00m\n\u001b[1;32m----> 4\u001b[0m dataset_path \u001b[38;5;241m=\u001b[39m os\u001b[38;5;241m.\u001b[39mlistdir(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mC:/Users/GB/Desktop/Taliqa/dataset/DS\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m      7\u001b[0m \u001b[38;5;28mprint\u001b[39m (dataset_path)  \u001b[38;5;66;03m#what kinds of classes are in this dataset\u001b[39;00m\n\u001b[0;32m      9\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mTypes of classes labels found: \u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28mlen\u001b[39m(dataset_path))\n",
      "\u001b[1;31mNameError\u001b[0m: name 'os' is not defined"
     ]
    }
   ],
   "source": [
    "# num_classes = 2\n",
    "# input_shape = (224, 224, 3)\n",
    "\n",
    "dataset_path = os.listdir('C:/Users/GB/Desktop/Taliqa/dataset/DS')\n",
    "\n",
    "\n",
    "print (dataset_path)  #what kinds of classes are in this dataset\n",
    "\n",
    "print(\"Types of classes labels found: \", len(dataset_path))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "967ec603",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'dataset_path' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[6], line 4\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mpandas\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mpd\u001b[39;00m \n\u001b[0;32m      2\u001b[0m class_labels \u001b[38;5;241m=\u001b[39m []\n\u001b[1;32m----> 4\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m item \u001b[38;5;129;01min\u001b[39;00m dataset_path:\n\u001b[0;32m      5\u001b[0m  \u001b[38;5;66;03m# Get all the file names\u001b[39;00m\n\u001b[0;32m      6\u001b[0m  all_classes \u001b[38;5;241m=\u001b[39m os\u001b[38;5;241m.\u001b[39mlistdir(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mC:/Users/GB/Desktop/Taliqa/dataset/DS/\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;241m+\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m/\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;241m+\u001b[39mitem)\n\u001b[0;32m      7\u001b[0m  \u001b[38;5;28mprint\u001b[39m(all_classes)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'dataset_path' is not defined"
     ]
    }
   ],
   "source": [
    "import pandas as pd \n",
    "class_labels = []\n",
    "\n",
    "for item in dataset_path:\n",
    " # Get all the file names\n",
    " all_classes = os.listdir('C:/Users/GB/Desktop/Taliqa/dataset/DS/' + '/' +item)\n",
    " print(all_classes)\n",
    "\n",
    "# Add them to the list\n",
    " for room in all_classes:\n",
    "    class_labels.append((item, str('dataset_path' + '/' +item) + '/' + room))\n",
    "    print(class_labels[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "12acd204",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Labels                                   image\n",
      "0  Normal     dataset_path/Normal/Normal- (1).jpg\n",
      "1  Normal    dataset_path/Normal/Normal- (10).jpg\n",
      "2  Normal   dataset_path/Normal/Normal- (100).jpg\n",
      "3  Normal  dataset_path/Normal/Normal- (1000).jpg\n",
      "4  Normal  dataset_path/Normal/Normal- (1001).jpg\n",
      "     Labels                                image\n",
      "9279  STONE  dataset_path/STONE/STONE- (995).jpg\n",
      "9280  STONE  dataset_path/STONE/STONE- (996).jpg\n",
      "9281  STONE  dataset_path/STONE/STONE- (997).jpg\n",
      "9282  STONE  dataset_path/STONE/STONE- (998).jpg\n",
      "9283  STONE  dataset_path/STONE/STONE- (999).jpg\n"
     ]
    }
   ],
   "source": [
    "# # Build a dataframe        \n",
    "df = pd.DataFrame(data=class_labels, columns=['Labels', 'image'])\n",
    "print(df.head())\n",
    "print(df.tail())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "80e33d86",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of images in the dataset:  9284\n",
      "Normal    5905\n",
      "STONE     3379\n",
      "Name: Labels, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Let's check how many samples for each category are present\n",
    "print(\"Total number of images in the dataset: \", len(df))\n",
    "\n",
    "label_count = df['Labels'].value_counts()\n",
    "print(label_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e394ee3c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import cv2\n",
    "from PIL import Image\n",
    "path = 'C:/Users/GB/Desktop/Taliqa/dataset/DS/'\n",
    "dataset_path = os.listdir('C:/Users/GB/Desktop/Taliqa/dataset/DS/')\n",
    "\n",
    "im_size = 224\n",
    "images = []\n",
    "labels = []\n",
    "\n",
    "for i in dataset_path:\n",
    "    data_path = path + str(i)  \n",
    "    filenames = [i for i in os.listdir(data_path) ]\n",
    "   \n",
    "    for f in filenames:\n",
    "        img = cv2.imread(data_path + '/' + f)\n",
    "        img = cv2.resize(img, (im_size, im_size))\n",
    "        #patches = extract_patches_2d(img, patch_size(224, 224), max_patches = 10)\n",
    "        #img = np.array(img)\n",
    "        images.append(img)\n",
    "        labels.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0df7a4e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "images = np.array(images)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1b0827a8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(9284, 224, 224, 3)\n"
     ]
    }
   ],
   "source": [
    "print(images.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "13dfee56",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Normal' 'Normal' 'Normal' ... 'STONE' 'STONE' 'STONE']\n",
      "[0 0 0 ... 1 1 1]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder , OneHotEncoder\n",
    "y=df['Labels'].values\n",
    "print(y)\n",
    "\n",
    "y_labelencoder = LabelEncoder ()\n",
    "y = y_labelencoder.fit_transform (y)\n",
    "print (y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eb549a4e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1. 0.]\n",
      " [1. 0.]\n",
      " [1. 0.]\n",
      " [1. 0.]\n",
      " [1. 0.]]\n",
      "[[1. 0.]\n",
      " [1. 0.]\n",
      " [1. 0.]\n",
      " ...\n",
      " [0. 1.]\n",
      " [0. 1.]\n",
      " [0. 1.]]\n"
     ]
    }
   ],
   "source": [
    "y=y.reshape(-1,1)\n",
    "\n",
    "from sklearn.compose import ColumnTransformer\n",
    "ct = ColumnTransformer([('my_ohe', OneHotEncoder(), [0])], remainder='passthrough')\n",
    "Y = ct.fit_transform(y) #.toarray()\n",
    "print(Y[:5])\n",
    "print(Y[35:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "83c39141",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9284"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y=(Y-1)\n",
    "len(Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7d05526a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.utils import shuffle\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n",
    "images, Y = shuffle(images, Y, random_state=1)\n",
    "\n",
    "\n",
    "train_x, test_x, train_y, test_y = train_test_split(images, Y, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1223a407",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: (7427, 224, 224, 3) - y_train shape: (7427, 2)\n",
      "x_test shape: (1857, 224, 224, 3) - y_test shape: (1857, 2)\n"
     ]
    }
   ],
   "source": [
    "print(f\"x_train shape: {train_x.shape} - y_train shape: {train_y.shape}\")\n",
    "print(f\"x_test shape: {test_x.shape} - y_test shape: {test_y.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "47a43762",
   "metadata": {},
   "outputs": [],
   "source": [
    "BUFFER_SIZE = 224\n",
    "BATCH_SIZE = 256\n",
    "\n",
    "# AUGMENTATION\n",
    "IMAGE_SIZE = 224\n",
    "PATCH_SIZE = 64\n",
    "NUM_PATCHES = (IMAGE_SIZE // PATCH_SIZE) ** 2\n",
    "\n",
    "# OPTIMIZER\n",
    "LEARNING_RATE = 0.001\n",
    "WEIGHT_DECAY = 0.0001\n",
    "\n",
    "# TRAINING\n",
    "EPOCHS = 50\n",
    "\n",
    "# ARCHITECTURE\n",
    "LAYER_NORM_EPS = 1e-6\n",
    "TRANSFORMER_LAYERS = 8\n",
    "PROJECTION_DIM = 64\n",
    "NUM_HEADS = 4\n",
    "TRANSFORMER_UNITS = [\n",
    "    PROJECTION_DIM * 2,\n",
    "    PROJECTION_DIM,\n",
    "]\n",
    "MLP_HEAD_UNITS = [2048, 1024]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0332ff8d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "data_augmentation =keras.Sequential(\n",
    "    [\n",
    "        keras.layers.Normalization(),\n",
    "        keras.layers.Resizing(IMAGE_SIZE, IMAGE_SIZE),\n",
    "        keras.layers.RandomFlip(\"horizontal\"),\n",
    "        keras.layers.RandomRotation(factor=0.02),\n",
    "        keras.layers.RandomZoom(height_factor=0.2, width_factor=0.2),\n",
    "    ],\n",
    "    name=\"data_augmentation\",\n",
    ")\n",
    "# Compute the mean and the variance of the training data for normalization.\n",
    "data_augmentation.layers[0].adapt(train_x)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80035f20",
   "metadata": {},
   "source": [
    "# Implement Shifted Patch Tokenization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dcc2393c",
   "metadata": {},
   "outputs": [],
   "source": [
    "class ShiftedPatchTokenization(keras.layers.Layer):\n",
    "    def __init__(\n",
    "        self,\n",
    "        image_size=IMAGE_SIZE,\n",
    "        patch_size=PATCH_SIZE,\n",
    "        num_patches=NUM_PATCHES,\n",
    "        projection_dim=PROJECTION_DIM,\n",
    "        vanilla=False,\n",
    "        **kwargs,\n",
    "    ):\n",
    "        super().__init__(**kwargs)\n",
    "        self.vanilla = vanilla  # Flag to swtich to vanilla patch extractor\n",
    "        self.image_size = image_size\n",
    "        self.patch_size = patch_size\n",
    "        self.half_patch = patch_size // 2\n",
    "        self.flatten_patches = keras.layers.Reshape((num_patches, -1))\n",
    "        self.projection = keras.layers.Dense(units=projection_dim)\n",
    "        self.layer_norm = keras.layers.LayerNormalization(epsilon=LAYER_NORM_EPS)\n",
    "\n",
    "    def crop_shift_pad(self, images, mode):\n",
    "        # Build the diagonally shifted images\n",
    "        if mode == \"left-up\":\n",
    "            crop_height = self.half_patch\n",
    "            crop_width = self.half_patch\n",
    "            shift_height = 0\n",
    "            shift_width = 0\n",
    "        elif mode == \"left-down\":\n",
    "            crop_height = 0\n",
    "            crop_width = self.half_patch\n",
    "            shift_height = self.half_patch\n",
    "            shift_width = 0\n",
    "        elif mode == \"right-up\":\n",
    "            crop_height = self.half_patch\n",
    "            crop_width = 0\n",
    "            shift_height = 0\n",
    "            shift_width = self.half_patch\n",
    "        else:\n",
    "            crop_height = 0\n",
    "            crop_width = 0\n",
    "            shift_height = self.half_patch\n",
    "            shift_width = self.half_patch\n",
    "\n",
    "        # Crop the shifted images and pad them\n",
    "        crop = tf.image.crop_to_bounding_box(\n",
    "            images,\n",
    "            offset_height=crop_height,\n",
    "            offset_width=crop_width,\n",
    "            target_height=self.image_size - self.half_patch,\n",
    "            target_width=self.image_size - self.half_patch,\n",
    "        )\n",
    "        shift_pad = tf.image.pad_to_bounding_box(\n",
    "            crop,\n",
    "            offset_height=shift_height,\n",
    "            offset_width=shift_width,\n",
    "            target_height=self.image_size,\n",
    "            target_width=self.image_size,\n",
    "        )\n",
    "        return shift_pad\n",
    "\n",
    "    def call(self, images):\n",
    "        if not self.vanilla:\n",
    "            # Concat the shifted images with the original image\n",
    "            images = tf.concat(\n",
    "                [\n",
    "                    images,\n",
    "                    self.crop_shift_pad(images, mode=\"left-up\"),\n",
    "                    self.crop_shift_pad(images, mode=\"left-down\"),\n",
    "                    self.crop_shift_pad(images, mode=\"right-up\"),\n",
    "                    self.crop_shift_pad(images, mode=\"right-down\"),\n",
    "                ],\n",
    "                axis=-1,\n",
    "            )\n",
    "        # Patchify the images and flatten it\n",
    "        patches = tf.image.extract_patches(\n",
    "            images=images,\n",
    "            sizes=[1, self.patch_size, self.patch_size, 1],\n",
    "            strides=[1, self.patch_size, self.patch_size, 1],\n",
    "            rates=[1, 1, 1, 1],\n",
    "            padding=\"VALID\",\n",
    "        )\n",
    "        flat_patches = self.flatten_patches(patches)\n",
    "        if not self.vanilla:\n",
    "            # Layer normalize the flat patches and linearly project it\n",
    "            tokens = self.layer_norm(flat_patches)\n",
    "            tokens = self.projection(tokens)\n",
    "        else:\n",
    "            # Linearly project the flat patches\n",
    "            tokens = self.projection(flat_patches)\n",
    "        return (tokens, patches)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "97a8d747",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "4e8c2566",
   "metadata": {},
   "source": [
    "# Visualize the patches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8f1720c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ORIGINAL\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LEFT-UP\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LEFT-DOWN\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RIGHT-UP\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RIGHT-DOWN\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get a random image from the training dataset\n",
    "# and resize the image\n",
    "image = train_x[np.random.choice(range(train_x.shape[0]))]\n",
    "resized_image = tf.image.resize(\n",
    "    tf.convert_to_tensor([image]), size=(IMAGE_SIZE, IMAGE_SIZE)\n",
    ")\n",
    "\n",
    "# Vanilla patch maker: This takes an image and divides into\n",
    "# patches as in the original ViT paper\n",
    "(token, patch) = ShiftedPatchTokenization(vanilla=True)(resized_image / 255.0)\n",
    "(token, patch) = (token[0], patch[0])\n",
    "n = patch.shape[0]\n",
    "count = 1\n",
    "plt.figure(figsize=(4, 4))\n",
    "for row in range(n):\n",
    "    for col in range(n):\n",
    "        plt.subplot(n, n, count)\n",
    "        count = count + 1\n",
    "        image = tf.reshape(patch[row][col], (PATCH_SIZE, PATCH_SIZE, 3))\n",
    "        plt.imshow(image)\n",
    "        plt.axis(\"off\")\n",
    "plt.show()\n",
    "\n",
    "# Shifted Patch Tokenization: This layer takes the image, shifts it\n",
    "# diagonally and then extracts patches from the concatinated images\n",
    "(token, patch) = ShiftedPatchTokenization(vanilla=False)(resized_image / 255.0)\n",
    "(token, patch) = (token[0], patch[0])\n",
    "n = patch.shape[0]\n",
    "shifted_images = [\"ORIGINAL\", \"LEFT-UP\", \"LEFT-DOWN\", \"RIGHT-UP\", \"RIGHT-DOWN\"]\n",
    "for index, name in enumerate(shifted_images):\n",
    "    print(name)\n",
    "    count = 1\n",
    "    plt.figure(figsize=(4, 4))\n",
    "    for row in range(n):\n",
    "        for col in range(n):\n",
    "            plt.subplot(n, n, count)\n",
    "            count = count + 1\n",
    "            image = tf.reshape(patch[row][col], (PATCH_SIZE, PATCH_SIZE, 5 * 3))\n",
    "            plt.imshow(image[..., 3 * index : 3 * index + 3])\n",
    "            plt.axis(\"off\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "119bf360",
   "metadata": {},
   "source": [
    "# Implement the patch encoding layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "10d0f48d",
   "metadata": {},
   "outputs": [],
   "source": [
    "class PatchEncoder(layers.Layer):\n",
    "    def __init__(\n",
    "        self, num_patches=NUM_PATCHES, projection_dim=PROJECTION_DIM, **kwargs\n",
    "    ):\n",
    "        super().__init__(**kwargs)\n",
    "        self.num_patches = num_patches\n",
    "        self.position_embedding = layers.Embedding(\n",
    "            input_dim=num_patches, output_dim=projection_dim\n",
    "        )\n",
    "        self.positions = tf.range(start=0, limit=self.num_patches, delta=1)\n",
    "\n",
    "    def call(self, encoded_patches):\n",
    "        encoded_positions = self.position_embedding(self.positions)\n",
    "        encoded_patches = encoded_patches + encoded_positions\n",
    "        return encoded_patches"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eea8284f",
   "metadata": {},
   "source": [
    "# Implement Locality Self Attention"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "89ccd806",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MultiHeadAttentionLSA(tf.keras.layers.MultiHeadAttention):\n",
    "    def __init__(self, **kwargs):\n",
    "        super().__init__(**kwargs)\n",
    "        # The trainable temperature term. The initial value is\n",
    "        # the square root of the key dimension.\n",
    "        self.tau = tf.Variable(math.sqrt(float(self._key_dim)), trainable=True)\n",
    "\n",
    "    def _compute_attention(self, query, key, value, attention_mask=None, training=None):\n",
    "        query = tf.multiply(query, 1.0 / self.tau)\n",
    "        attention_scores = tf.einsum(self._dot_product_equation, key, query)\n",
    "        attention_scores = self._masked_softmax(attention_scores, attention_mask)\n",
    "        attention_scores_dropout = self._dropout_layer(\n",
    "            attention_scores, training=training\n",
    "        )\n",
    "        attention_output = tf.einsum(\n",
    "            self._combine_equation, attention_scores_dropout, value\n",
    "        )\n",
    "        return attention_output, attention_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ff30bbb9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def mlp(x, hidden_units, dropout_rate):\n",
    "    for units in hidden_units:\n",
    "        x = layers.Dense(units, activation=tf.nn.gelu)(x)\n",
    "        x = layers.Dropout(dropout_rate)(x)\n",
    "    return x\n",
    "\n",
    "# Build the diagonal attention mask\n",
    "diag_attn_mask = 1 - tf.eye(NUM_PATCHES)\n",
    "diag_attn_mask = tf.cast([diag_attn_mask], dtype=tf.int8)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a0f3d1d",
   "metadata": {},
   "source": [
    "# Build the ViT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6229a350",
   "metadata": {},
   "outputs": [],
   "source": [
    "INPUT_SHAPE = (224, 224, 3)\n",
    "NUM_CLASSES = 2\n",
    "def create_vit_classifier(vanilla=False):\n",
    "    inputs = layers.Input(shape=INPUT_SHAPE)\n",
    "    # Augment data.\n",
    "    augmented = data_augmentation(inputs)\n",
    "    # Create patches.\n",
    "    (tokens, _) = ShiftedPatchTokenization(vanilla=vanilla)(augmented)\n",
    "    # Encode patches.\n",
    "    encoded_patches = PatchEncoder()(tokens)\n",
    "\n",
    "    # Create multiple layers of the Transformer block.\n",
    "    for _ in range(TRANSFORMER_LAYERS):\n",
    "        # Layer normalization 1.\n",
    "        x1 = layers.LayerNormalization(epsilon=1e-6)(encoded_patches)\n",
    "        # Create a multi-head attention layer.\n",
    "        if not vanilla:\n",
    "            attention_output = MultiHeadAttentionLSA(\n",
    "                num_heads=NUM_HEADS, key_dim=PROJECTION_DIM, dropout=0.1\n",
    "            )(x1, x1, attention_mask=diag_attn_mask)\n",
    "        else:\n",
    "            attention_output = layers.MultiHeadAttention(\n",
    "                num_heads=NUM_HEADS, key_dim=PROJECTION_DIM, dropout=0.1\n",
    "            )(x1, x1)\n",
    "        # Skip connection 1.\n",
    "        x2 = layers.Add()([attention_output, encoded_patches])\n",
    "        # Layer normalization 2.\n",
    "        x3 = layers.LayerNormalization(epsilon=1e-6)(x2)\n",
    "        # MLP.\n",
    "        x3 = mlp(x3, hidden_units=TRANSFORMER_UNITS, dropout_rate=0.1)\n",
    "        # Skip connection 2.\n",
    "        encoded_patches = layers.Add()([x3, x2])\n",
    "\n",
    "    # Create a [batch_size, projection_dim] tensor.\n",
    "    representation = layers.LayerNormalization(epsilon=1e-6)(encoded_patches)\n",
    "    representation = layers.Flatten()(representation)\n",
    "    representation = layers.Dropout(0.5)(representation)\n",
    "    # Add MLP.\n",
    "    features = mlp(representation, hidden_units=MLP_HEAD_UNITS, dropout_rate=0.5)\n",
    "    # Classify outputs.\n",
    "    logits = layers.Dense(NUM_CLASSES)(features)\n",
    "    # Create the Keras model.\n",
    "    model = keras.Model(inputs=inputs, outputs=logits)\n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "009b4459",
   "metadata": {},
   "source": [
    "# Compile, train, and evaluate the mode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6923cdec",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "27/27 [==============================] - 67s 2s/step - loss: 2.4165 - accuracy: 0.5741 - val_loss: 0.2090 - val_accuracy: 0.7510\n",
      "Epoch 2/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.2860 - accuracy: 0.6587 - val_loss: 0.1863 - val_accuracy: 0.7604\n",
      "Epoch 3/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.2511 - accuracy: 0.6764 - val_loss: 0.1660 - val_accuracy: 0.7793\n",
      "Epoch 4/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.2366 - accuracy: 0.6991 - val_loss: 0.1664 - val_accuracy: 0.7820\n",
      "Epoch 5/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.2268 - accuracy: 0.7124 - val_loss: 0.1609 - val_accuracy: 0.7995\n",
      "Epoch 6/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.2168 - accuracy: 0.7208 - val_loss: 0.1697 - val_accuracy: 0.7981\n",
      "Epoch 7/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.2065 - accuracy: 0.7379 - val_loss: 0.1674 - val_accuracy: 0.8102\n",
      "Epoch 8/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.2006 - accuracy: 0.7400 - val_loss: 0.1829 - val_accuracy: 0.8102\n",
      "Epoch 9/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1953 - accuracy: 0.7546 - val_loss: 0.1688 - val_accuracy: 0.8183\n",
      "Epoch 10/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1875 - accuracy: 0.7579 - val_loss: 0.1688 - val_accuracy: 0.8116\n",
      "Epoch 11/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1811 - accuracy: 0.7697 - val_loss: 0.1928 - val_accuracy: 0.8210\n",
      "Epoch 12/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1749 - accuracy: 0.7787 - val_loss: 0.1935 - val_accuracy: 0.8116\n",
      "Epoch 13/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1702 - accuracy: 0.7828 - val_loss: 0.1590 - val_accuracy: 0.8358\n",
      "Epoch 14/100\n",
      "27/27 [==============================] - 44s 2s/step - loss: 0.1618 - accuracy: 0.7974 - val_loss: 0.1363 - val_accuracy: 0.8654\n",
      "Epoch 15/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1552 - accuracy: 0.8060 - val_loss: 0.1387 - val_accuracy: 0.8708\n",
      "Epoch 16/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1475 - accuracy: 0.8101 - val_loss: 0.1548 - val_accuracy: 0.8694\n",
      "Epoch 17/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1402 - accuracy: 0.8235 - val_loss: 0.1373 - val_accuracy: 0.8708\n",
      "Epoch 18/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1354 - accuracy: 0.8302 - val_loss: 0.1417 - val_accuracy: 0.8896\n",
      "Epoch 19/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1342 - accuracy: 0.8309 - val_loss: 0.1591 - val_accuracy: 0.8748\n",
      "Epoch 20/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1312 - accuracy: 0.8302 - val_loss: 0.1336 - val_accuracy: 0.8843\n",
      "Epoch 21/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1252 - accuracy: 0.8417 - val_loss: 0.1286 - val_accuracy: 0.8816\n",
      "Epoch 22/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1176 - accuracy: 0.8532 - val_loss: 0.1125 - val_accuracy: 0.8964\n",
      "Epoch 23/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1144 - accuracy: 0.8576 - val_loss: 0.1017 - val_accuracy: 0.9071\n",
      "Epoch 24/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1133 - accuracy: 0.8615 - val_loss: 0.1030 - val_accuracy: 0.8991\n",
      "Epoch 25/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1117 - accuracy: 0.8627 - val_loss: 0.1005 - val_accuracy: 0.9112\n",
      "Epoch 26/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1068 - accuracy: 0.8668 - val_loss: 0.0972 - val_accuracy: 0.9192\n",
      "Epoch 27/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1013 - accuracy: 0.8760 - val_loss: 0.1075 - val_accuracy: 0.9219\n",
      "Epoch 28/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0974 - accuracy: 0.8838 - val_loss: 0.0985 - val_accuracy: 0.9192\n",
      "Epoch 29/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0960 - accuracy: 0.8821 - val_loss: 0.1096 - val_accuracy: 0.9166\n",
      "Epoch 30/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0940 - accuracy: 0.8823 - val_loss: 0.0976 - val_accuracy: 0.9166\n",
      "Epoch 31/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0957 - accuracy: 0.8833 - val_loss: 0.1199 - val_accuracy: 0.9125\n",
      "Epoch 32/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.1047 - accuracy: 0.8667 - val_loss: 0.1033 - val_accuracy: 0.9125\n",
      "Epoch 33/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0917 - accuracy: 0.8866 - val_loss: 0.0836 - val_accuracy: 0.9341\n",
      "Epoch 34/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0876 - accuracy: 0.8917 - val_loss: 0.0871 - val_accuracy: 0.9354\n",
      "Epoch 35/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0832 - accuracy: 0.8984 - val_loss: 0.0854 - val_accuracy: 0.9341\n",
      "Epoch 36/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0814 - accuracy: 0.8993 - val_loss: 0.0844 - val_accuracy: 0.9327\n",
      "Epoch 37/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0828 - accuracy: 0.8953 - val_loss: 0.0812 - val_accuracy: 0.9341\n",
      "Epoch 38/100\n",
      "27/27 [==============================] - 16089s 619s/step - loss: 0.0812 - accuracy: 0.9002 - val_loss: 0.0789 - val_accuracy: 0.9354\n",
      "Epoch 39/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0785 - accuracy: 0.9025 - val_loss: 0.0795 - val_accuracy: 0.9381\n",
      "Epoch 40/100\n",
      "27/27 [==============================] - 52s 2s/step - loss: 0.0746 - accuracy: 0.9065 - val_loss: 0.0794 - val_accuracy: 0.9354\n",
      "Epoch 41/100\n",
      "27/27 [==============================] - 49s 2s/step - loss: 0.0739 - accuracy: 0.9080 - val_loss: 0.0753 - val_accuracy: 0.9381\n",
      "Epoch 42/100\n",
      "27/27 [==============================] - 49s 2s/step - loss: 0.0734 - accuracy: 0.9083 - val_loss: 0.0797 - val_accuracy: 0.9367\n",
      "Epoch 43/100\n",
      "27/27 [==============================] - 55s 2s/step - loss: 0.0707 - accuracy: 0.9092 - val_loss: 0.0731 - val_accuracy: 0.9314\n",
      "Epoch 44/100\n",
      "27/27 [==============================] - 51s 2s/step - loss: 0.0710 - accuracy: 0.9089 - val_loss: 0.0698 - val_accuracy: 0.9367\n",
      "Epoch 45/100\n",
      "27/27 [==============================] - 49s 2s/step - loss: 0.0720 - accuracy: 0.9102 - val_loss: 0.0739 - val_accuracy: 0.9354\n",
      "Epoch 46/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0708 - accuracy: 0.9110 - val_loss: 0.0650 - val_accuracy: 0.9367\n",
      "Epoch 47/100\n",
      "27/27 [==============================] - 50s 2s/step - loss: 0.0686 - accuracy: 0.9152 - val_loss: 0.0717 - val_accuracy: 0.9394\n",
      "Epoch 48/100\n",
      "27/27 [==============================] - 50s 2s/step - loss: 0.0657 - accuracy: 0.9198 - val_loss: 0.0750 - val_accuracy: 0.9435\n",
      "Epoch 49/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0675 - accuracy: 0.9138 - val_loss: 0.0665 - val_accuracy: 0.9462\n",
      "Epoch 50/100\n",
      "27/27 [==============================] - 50s 2s/step - loss: 0.0673 - accuracy: 0.9158 - val_loss: 0.0594 - val_accuracy: 0.9381\n",
      "Epoch 51/100\n",
      "27/27 [==============================] - 49s 2s/step - loss: 0.0654 - accuracy: 0.9195 - val_loss: 0.0638 - val_accuracy: 0.9408\n",
      "Epoch 52/100\n",
      "27/27 [==============================] - 51s 2s/step - loss: 0.0643 - accuracy: 0.9185 - val_loss: 0.0646 - val_accuracy: 0.9408\n",
      "Epoch 53/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0629 - accuracy: 0.9215 - val_loss: 0.0720 - val_accuracy: 0.9381\n",
      "Epoch 54/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0642 - accuracy: 0.9186 - val_loss: 0.0656 - val_accuracy: 0.9435\n",
      "Epoch 55/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0676 - accuracy: 0.9146 - val_loss: 0.0694 - val_accuracy: 0.9408\n",
      "Epoch 56/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0645 - accuracy: 0.9164 - val_loss: 0.0634 - val_accuracy: 0.9462\n",
      "Epoch 57/100\n",
      "27/27 [==============================] - 52s 2s/step - loss: 0.0618 - accuracy: 0.9185 - val_loss: 0.0563 - val_accuracy: 0.9462\n",
      "Epoch 58/100\n",
      "27/27 [==============================] - 51s 2s/step - loss: 0.0619 - accuracy: 0.9180 - val_loss: 0.0602 - val_accuracy: 0.9448\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 59/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0601 - accuracy: 0.9241 - val_loss: 0.0555 - val_accuracy: 0.9448\n",
      "Epoch 60/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0585 - accuracy: 0.9258 - val_loss: 0.0575 - val_accuracy: 0.9435\n",
      "Epoch 61/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0644 - accuracy: 0.9189 - val_loss: 0.0615 - val_accuracy: 0.9408\n",
      "Epoch 62/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0604 - accuracy: 0.9232 - val_loss: 0.0511 - val_accuracy: 0.9542\n",
      "Epoch 63/100\n",
      "27/27 [==============================] - 51s 2s/step - loss: 0.0618 - accuracy: 0.9204 - val_loss: 0.0500 - val_accuracy: 0.9462\n",
      "Epoch 64/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0583 - accuracy: 0.9255 - val_loss: 0.0520 - val_accuracy: 0.9502\n",
      "Epoch 65/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0569 - accuracy: 0.9295 - val_loss: 0.0537 - val_accuracy: 0.9489\n",
      "Epoch 66/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0574 - accuracy: 0.9268 - val_loss: 0.0489 - val_accuracy: 0.9475\n",
      "Epoch 67/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0570 - accuracy: 0.9255 - val_loss: 0.0571 - val_accuracy: 0.9542\n",
      "Epoch 68/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0563 - accuracy: 0.9253 - val_loss: 0.0545 - val_accuracy: 0.9448\n",
      "Epoch 69/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0602 - accuracy: 0.9213 - val_loss: 0.0523 - val_accuracy: 0.9489\n",
      "Epoch 70/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0573 - accuracy: 0.9262 - val_loss: 0.0433 - val_accuracy: 0.9542\n",
      "Epoch 71/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0549 - accuracy: 0.9289 - val_loss: 0.0472 - val_accuracy: 0.9596\n",
      "Epoch 72/100\n",
      "27/27 [==============================] - 45s 2s/step - loss: 0.0517 - accuracy: 0.9354 - val_loss: 0.0469 - val_accuracy: 0.9542\n",
      "Epoch 73/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0513 - accuracy: 0.9333 - val_loss: 0.0495 - val_accuracy: 0.9542\n",
      "Epoch 74/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0525 - accuracy: 0.9343 - val_loss: 0.0463 - val_accuracy: 0.9515\n",
      "Epoch 75/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0533 - accuracy: 0.9292 - val_loss: 0.0471 - val_accuracy: 0.9529\n",
      "Epoch 76/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0528 - accuracy: 0.9334 - val_loss: 0.0416 - val_accuracy: 0.9529\n",
      "Epoch 77/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0516 - accuracy: 0.9331 - val_loss: 0.0418 - val_accuracy: 0.9556\n",
      "Epoch 78/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0513 - accuracy: 0.9334 - val_loss: 0.0434 - val_accuracy: 0.9529\n",
      "Epoch 79/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0515 - accuracy: 0.9307 - val_loss: 0.0449 - val_accuracy: 0.9529\n",
      "Epoch 80/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0512 - accuracy: 0.9312 - val_loss: 0.0451 - val_accuracy: 0.9489\n",
      "Epoch 81/100\n",
      "27/27 [==============================] - 50s 2s/step - loss: 0.0525 - accuracy: 0.9310 - val_loss: 0.0404 - val_accuracy: 0.9502\n",
      "Epoch 82/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0511 - accuracy: 0.9319 - val_loss: 0.0405 - val_accuracy: 0.9515\n",
      "Epoch 83/100\n",
      "27/27 [==============================] - 52s 2s/step - loss: 0.0541 - accuracy: 0.9277 - val_loss: 0.0413 - val_accuracy: 0.9556\n",
      "Epoch 84/100\n",
      "27/27 [==============================] - 50s 2s/step - loss: 0.0502 - accuracy: 0.9333 - val_loss: 0.0415 - val_accuracy: 0.9462\n",
      "Epoch 85/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0511 - accuracy: 0.9321 - val_loss: 0.0412 - val_accuracy: 0.9583\n",
      "Epoch 86/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0502 - accuracy: 0.9325 - val_loss: 0.0402 - val_accuracy: 0.9569\n",
      "Epoch 87/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0502 - accuracy: 0.9334 - val_loss: 0.0409 - val_accuracy: 0.9556\n",
      "Epoch 88/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0517 - accuracy: 0.9309 - val_loss: 0.0412 - val_accuracy: 0.9569\n",
      "Epoch 89/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0520 - accuracy: 0.9312 - val_loss: 0.0382 - val_accuracy: 0.9610\n",
      "Epoch 90/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0519 - accuracy: 0.9315 - val_loss: 0.0386 - val_accuracy: 0.9556\n",
      "Epoch 91/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0512 - accuracy: 0.9334 - val_loss: 0.0461 - val_accuracy: 0.9515\n",
      "Epoch 92/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0549 - accuracy: 0.9261 - val_loss: 0.0386 - val_accuracy: 0.9569\n",
      "Epoch 93/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0495 - accuracy: 0.9342 - val_loss: 0.0378 - val_accuracy: 0.9610\n",
      "Epoch 94/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0486 - accuracy: 0.9370 - val_loss: 0.0373 - val_accuracy: 0.9596\n",
      "Epoch 95/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0486 - accuracy: 0.9372 - val_loss: 0.0359 - val_accuracy: 0.9542\n",
      "Epoch 96/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0499 - accuracy: 0.9325 - val_loss: 0.0389 - val_accuracy: 0.9556\n",
      "Epoch 97/100\n",
      "27/27 [==============================] - 48s 2s/step - loss: 0.0487 - accuracy: 0.9352 - val_loss: 0.0389 - val_accuracy: 0.9569\n",
      "Epoch 98/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0469 - accuracy: 0.9385 - val_loss: 0.0390 - val_accuracy: 0.9556\n",
      "Epoch 99/100\n",
      "27/27 [==============================] - 47s 2s/step - loss: 0.0473 - accuracy: 0.9378 - val_loss: 0.0393 - val_accuracy: 0.9542\n",
      "Epoch 100/100\n",
      "27/27 [==============================] - 46s 2s/step - loss: 0.0493 - accuracy: 0.9379 - val_loss: 0.0358 - val_accuracy: 0.9569\n"
     ]
    }
   ],
   "source": [
    "# Some code is taken from:\n",
    "# https://www.kaggle.com/ashusma/training-rfcx-tensorflow-tpu-effnet-b2.\n",
    "class WarmUpCosine(keras.optimizers.schedules.LearningRateSchedule):\n",
    "    def __init__(\n",
    "        self, learning_rate_base, total_steps, warmup_learning_rate, warmup_steps\n",
    "    ):\n",
    "        super(WarmUpCosine, self).__init__()\n",
    "\n",
    "        self.learning_rate_base = learning_rate_base\n",
    "        self.total_steps = total_steps\n",
    "        self.warmup_learning_rate = warmup_learning_rate\n",
    "        self.warmup_steps = warmup_steps\n",
    "        self.pi = tf.constant(np.pi)\n",
    "\n",
    "    def __call__(self, step):\n",
    "        if self.total_steps < self.warmup_steps:\n",
    "            raise ValueError(\"Total_steps must be larger or equal to warmup_steps.\")\n",
    "\n",
    "        cos_annealed_lr = tf.cos(\n",
    "            self.pi\n",
    "            * (tf.cast(step, tf.float32) - self.warmup_steps)\n",
    "            / float(self.total_steps - self.warmup_steps)\n",
    "        )\n",
    "        learning_rate = 0.5 * self.learning_rate_base * (1 + cos_annealed_lr)\n",
    "\n",
    "        if self.warmup_steps > 0:\n",
    "            if self.learning_rate_base < self.warmup_learning_rate:\n",
    "                raise ValueError(\n",
    "                    \"Learning_rate_base must be larger or equal to \"\n",
    "                    \"warmup_learning_rate.\"\n",
    "                )\n",
    "            slope = (\n",
    "                self.learning_rate_base - self.warmup_learning_rate\n",
    "            ) / self.warmup_steps\n",
    "            warmup_rate = slope * tf.cast(step, tf.float32) + self.warmup_learning_rate\n",
    "            learning_rate = tf.where(\n",
    "                step < self.warmup_steps, warmup_rate, learning_rate\n",
    "            )\n",
    "        return tf.where(\n",
    "            step > self.total_steps, 0.0, learning_rate, name=\"learning_rate\"\n",
    "        )\n",
    "\n",
    "\n",
    "def run_experiment(model):\n",
    "    total_steps = int((len(train_x) / BATCH_SIZE) * EPOCHS)\n",
    "    warmup_epoch_percentage = 0.10\n",
    "    warmup_steps = int(total_steps * warmup_epoch_percentage)\n",
    "    scheduled_lrs = WarmUpCosine(\n",
    "        learning_rate_base=LEARNING_RATE,\n",
    "        total_steps=total_steps,\n",
    "        warmup_learning_rate=0.001,\n",
    "        warmup_steps=warmup_steps,\n",
    "    )\n",
    "\n",
    "    optimizer = tfa.optimizers.AdamW(\n",
    "        learning_rate=LEARNING_RATE, weight_decay=WEIGHT_DECAY\n",
    "    )    \n",
    "    model.compile(\n",
    "        optimizer=optimizer,\n",
    "        #loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True)\n",
    "        #loss= tf.keras.losses.CategoricalCrossentropy(from_logits=True)\n",
    "        #loss=tf.keras.losses.Huber()\n",
    "        loss=tf.keras.losses.MeanSquaredError(),\n",
    "        metrics=['accuracy']\n",
    "            #keras.metrics.SparseCategoricalAccuracy(name=\"accuracy\")\n",
    "            #tf.keras.metrics.CategoricalAccuracy(h),\n",
    "            #keras.metrics.SparseTopKCategoricalAccuracy(5, name=\"top-5-accuracy\"),\n",
    "     #   ]\n",
    "    )\n",
    "    checkpoint_path = \"training_1/cp.ckpt\"\n",
    "    checkpoint_dir = os.path.dirname(checkpoint_path)\n",
    "\n",
    "# Create a callback that saves the model's weights\n",
    "    cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_path,save_weights_only=True,verbose=1)\n",
    "\n",
    "    history = model.fit(\n",
    "        x=train_x,\n",
    "        y=train_y,\n",
    "        batch_size=BATCH_SIZE,\n",
    "        epochs=100,\n",
    "        validation_split=0.1,\n",
    "    )\n",
    "\n",
    "\n",
    "    return history\n",
    "\n",
    "\n",
    "# Run experiments with the vanilla ViT\n",
    "vit = create_vit_classifier(vanilla=True)\n",
    "history = run_experiment(vit)\n",
    "\n",
    "# Run experiments with the Shifted Patch Tokenization and\n",
    "# Locality Self Attention modified ViT\n",
    "#vit_sl = create_vit_classifier(vanilla=False)\n",
    "#history = run_experiment(vit_sl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9505f6ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2880x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs = [i for i in range(100)]\n",
    "fig , ax = plt.subplots(1,2)\n",
    "train_acc = history.history['accuracy']\n",
    "train_loss = history.history['loss']\n",
    "val_acc = history.history['val_accuracy']\n",
    "val_loss = history.history['val_loss']\n",
    "fig.set_size_inches(40,8)\n",
    "\n",
    "ax[0].plot(epochs , train_acc , 'go-' , label = 'Training Accuracy')\n",
    "ax[0].plot(epochs , val_acc , 'ro-' , label = 'Validation Accuracy')\n",
    "ax[0].set_title('Training & Validation Accuracy')\n",
    "ax[0].legend()\n",
    "ax[0].set_xlabel(\"Epochs\")\n",
    "ax[0].set_ylabel(\"Accuracy\")\n",
    "\n",
    "ax[1].plot(epochs , train_loss , 'g-o' , label = 'Training Loss')\n",
    "ax[1].plot(epochs , val_loss , 'r-o' , label = 'Validation Loss')\n",
    "ax[1].set_title('Testing Accuracy & Loss')\n",
    "ax[1].legend()\n",
    "ax[1].set_xlabel(\"Epochs\")\n",
    "ax[1].set_ylabel(\"Training & Validation Loss\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c59c1568",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59/59 [==============================] - 8s 107ms/step\n"
     ]
    }
   ],
   "source": [
    "y_pred=vit.predict(test_x)\n",
    "y_pred=np.argmax(y_pred, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1f57d51a",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_y=np.argmax(test_y, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "23472ae2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1140   35]\n",
      " [  76  606]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      0.97      0.95      1175\n",
      "           1       0.95      0.89      0.92       682\n",
      "\n",
      "    accuracy                           0.94      1857\n",
      "   macro avg       0.94      0.93      0.93      1857\n",
      "weighted avg       0.94      0.94      0.94      1857\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "print(confusion_matrix(test_y,y_pred))\n",
    "print(classification_report(test_y,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8409be51",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Confusion Matrix')"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.metrics import confusion_matrix\n",
    "cm_LR=confusion_matrix(test_y, y_pred)\n",
    "cmd = ConfusionMatrixDisplay(cm_LR, display_labels=['Normal' ,'Stone'])\n",
    "cmd.plot()\n",
    "plt.title('Confusion Matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "01b85a9d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<function create_vit_classifier at 0x00000209D725C9D0>\n"
     ]
    }
   ],
   "source": [
    "print(create_vit_classifier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "07a7fb6a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Skipping full serialization of Keras layer <keras.layers.normalization.layer_normalization.LayerNormalization object at 0x0000020AA69DBA60>, because it is not built.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as reshape_2_layer_call_fn, reshape_2_layer_call_and_return_conditional_losses, dense_2_layer_call_fn, dense_2_layer_call_and_return_conditional_losses, embedding_layer_call_fn while saving (showing 5 of 102). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: C:/Users/GB/Desktop/dataset/model\\assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: C:/Users/GB/Desktop/dataset/model\\assets\n"
     ]
    }
   ],
   "source": [
    "vit.save('C:/Users/GB/Desktop/dataset/model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3c635941",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'keras' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Input \u001b[1;32mIn [6]\u001b[0m, in \u001b[0;36m<cell line: 1>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[0m \u001b[43mkeras\u001b[49m\u001b[38;5;241m.\u001b[39mmodels\u001b[38;5;241m.\u001b[39mload_model(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mC:/Users/GB/Desktop/dataset/model\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'keras' is not defined"
     ]
    }
   ],
   "source": [
    "keras.models.load_model(\"C:/Users/GB/Desktop/dataset/model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7c3a5f11",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'vit' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Input \u001b[1;32mIn [5]\u001b[0m, in \u001b[0;36m<cell line: 1>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[0m \u001b[43mvit\u001b[49m\u001b[38;5;241m.\u001b[39msave_weights(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m./checkpoints/my_checkpoint\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'vit' is not defined"
     ]
    }
   ],
   "source": [
    "vit.save_weights('./checkpoints/my_checkpoint')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f7b78e53",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: pyyaml in c:\\users\\gb\\anaconda3\\lib\\site-packages (6.0)\n",
      "Requirement already satisfied: h5py in c:\\users\\gb\\anaconda3\\lib\\site-packages (3.6.0)\n",
      "Requirement already satisfied: numpy>=1.14.5 in c:\\users\\gb\\anaconda3\\lib\\site-packages (from h5py) (1.21.5)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install pyyaml h5py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5471918a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Volume in drive C has no label.\n",
      " Volume Serial Number is 48F6-B219\n",
      "\n",
      " Directory of C:\\Users\\GB\\Desktop\\dataset\\model\n",
      "\n",
      "05/08/2022  13:00    <DIR>          .\n",
      "05/08/2022  13:00    <DIR>          ..\n",
      "04/08/2022  09:57    <DIR>          assets\n",
      "05/08/2022  13:00           143,839 keras_metadata.pb\n",
      "05/08/2022  13:00         2,840,663 saved_model.pb\n",
      "05/08/2022  13:00    <DIR>          variables\n",
      "               2 File(s)      2,984,502 bytes\n",
      "               4 Dir(s)  58,903,584,768 bytes free\n"
     ]
    }
   ],
   "source": [
    "ls \"C:/Users/GB/Desktop/dataset/model\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "92dc2487",
   "metadata": {},
   "outputs": [],
   "source": [
    "new_model = tf.keras.models.load_model('C:/Users/GB/Desktop/dataset/model/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b98b66fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " input_1 (InputLayer)           [(None, 224, 224, 3  0           []                               \n",
      "                                )]                                                                \n",
      "                                                                                                  \n",
      " data_augmentation (Sequential)  (None, 224, 224, 3)  7          ['input_1[0][0]']                \n",
      "                                                                                                  \n",
      " shifted_patch_tokenization_2 (  ((None, 9, 64),     786496      ['data_augmentation[0][0]']      \n",
      " ShiftedPatchTokenization)       (None, 3, 3, 12288                                               \n",
      "                                ))                                                                \n",
      "                                                                                                  \n",
      " patch_encoder (PatchEncoder)   (None, 9, 64)        576         ['shifted_patch_tokenization_2[0]\n",
      "                                                                 [0]']                            \n",
      "                                                                                                  \n",
      " layer_normalization_3 (LayerNo  (None, 9, 64)       128         ['patch_encoder[0][0]']          \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " multi_head_attention (MultiHea  (None, 9, 64)       66368       ['layer_normalization_3[0][0]',  \n",
      " dAttention)                                                      'layer_normalization_3[0][0]']  \n",
      "                                                                                                  \n",
      " add (Add)                      (None, 9, 64)        0           ['multi_head_attention[0][0]',   \n",
      "                                                                  'patch_encoder[0][0]']          \n",
      "                                                                                                  \n",
      " layer_normalization_4 (LayerNo  (None, 9, 64)       128         ['add[0][0]']                    \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " dense_3 (Dense)                (None, 9, 128)       8320        ['layer_normalization_4[0][0]']  \n",
      "                                                                                                  \n",
      " dropout (Dropout)              (None, 9, 128)       0           ['dense_3[0][0]']                \n",
      "                                                                                                  \n",
      " dense_4 (Dense)                (None, 9, 64)        8256        ['dropout[0][0]']                \n",
      "                                                                                                  \n",
      " dropout_1 (Dropout)            (None, 9, 64)        0           ['dense_4[0][0]']                \n",
      "                                                                                                  \n",
      " add_1 (Add)                    (None, 9, 64)        0           ['dropout_1[0][0]',              \n",
      "                                                                  'add[0][0]']                    \n",
      "                                                                                                  \n",
      " layer_normalization_5 (LayerNo  (None, 9, 64)       128         ['add_1[0][0]']                  \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " multi_head_attention_1 (MultiH  (None, 9, 64)       66368       ['layer_normalization_5[0][0]',  \n",
      " eadAttention)                                                    'layer_normalization_5[0][0]']  \n",
      "                                                                                                  \n",
      " add_2 (Add)                    (None, 9, 64)        0           ['multi_head_attention_1[0][0]', \n",
      "                                                                  'add_1[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_6 (LayerNo  (None, 9, 64)       128         ['add_2[0][0]']                  \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " dense_5 (Dense)                (None, 9, 128)       8320        ['layer_normalization_6[0][0]']  \n",
      "                                                                                                  \n",
      " dropout_2 (Dropout)            (None, 9, 128)       0           ['dense_5[0][0]']                \n",
      "                                                                                                  \n",
      " dense_6 (Dense)                (None, 9, 64)        8256        ['dropout_2[0][0]']              \n",
      "                                                                                                  \n",
      " dropout_3 (Dropout)            (None, 9, 64)        0           ['dense_6[0][0]']                \n",
      "                                                                                                  \n",
      " add_3 (Add)                    (None, 9, 64)        0           ['dropout_3[0][0]',              \n",
      "                                                                  'add_2[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_7 (LayerNo  (None, 9, 64)       128         ['add_3[0][0]']                  \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " multi_head_attention_2 (MultiH  (None, 9, 64)       66368       ['layer_normalization_7[0][0]',  \n",
      " eadAttention)                                                    'layer_normalization_7[0][0]']  \n",
      "                                                                                                  \n",
      " add_4 (Add)                    (None, 9, 64)        0           ['multi_head_attention_2[0][0]', \n",
      "                                                                  'add_3[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_8 (LayerNo  (None, 9, 64)       128         ['add_4[0][0]']                  \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " dense_7 (Dense)                (None, 9, 128)       8320        ['layer_normalization_8[0][0]']  \n",
      "                                                                                                  \n",
      " dropout_4 (Dropout)            (None, 9, 128)       0           ['dense_7[0][0]']                \n",
      "                                                                                                  \n",
      " dense_8 (Dense)                (None, 9, 64)        8256        ['dropout_4[0][0]']              \n",
      "                                                                                                  \n",
      " dropout_5 (Dropout)            (None, 9, 64)        0           ['dense_8[0][0]']                \n",
      "                                                                                                  \n",
      " add_5 (Add)                    (None, 9, 64)        0           ['dropout_5[0][0]',              \n",
      "                                                                  'add_4[0][0]']                  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " layer_normalization_9 (LayerNo  (None, 9, 64)       128         ['add_5[0][0]']                  \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " multi_head_attention_3 (MultiH  (None, 9, 64)       66368       ['layer_normalization_9[0][0]',  \n",
      " eadAttention)                                                    'layer_normalization_9[0][0]']  \n",
      "                                                                                                  \n",
      " add_6 (Add)                    (None, 9, 64)        0           ['multi_head_attention_3[0][0]', \n",
      "                                                                  'add_5[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_10 (LayerN  (None, 9, 64)       128         ['add_6[0][0]']                  \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " dense_9 (Dense)                (None, 9, 128)       8320        ['layer_normalization_10[0][0]'] \n",
      "                                                                                                  \n",
      " dropout_6 (Dropout)            (None, 9, 128)       0           ['dense_9[0][0]']                \n",
      "                                                                                                  \n",
      " dense_10 (Dense)               (None, 9, 64)        8256        ['dropout_6[0][0]']              \n",
      "                                                                                                  \n",
      " dropout_7 (Dropout)            (None, 9, 64)        0           ['dense_10[0][0]']               \n",
      "                                                                                                  \n",
      " add_7 (Add)                    (None, 9, 64)        0           ['dropout_7[0][0]',              \n",
      "                                                                  'add_6[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_11 (LayerN  (None, 9, 64)       128         ['add_7[0][0]']                  \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " multi_head_attention_4 (MultiH  (None, 9, 64)       66368       ['layer_normalization_11[0][0]', \n",
      " eadAttention)                                                    'layer_normalization_11[0][0]'] \n",
      "                                                                                                  \n",
      " add_8 (Add)                    (None, 9, 64)        0           ['multi_head_attention_4[0][0]', \n",
      "                                                                  'add_7[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_12 (LayerN  (None, 9, 64)       128         ['add_8[0][0]']                  \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " dense_11 (Dense)               (None, 9, 128)       8320        ['layer_normalization_12[0][0]'] \n",
      "                                                                                                  \n",
      " dropout_8 (Dropout)            (None, 9, 128)       0           ['dense_11[0][0]']               \n",
      "                                                                                                  \n",
      " dense_12 (Dense)               (None, 9, 64)        8256        ['dropout_8[0][0]']              \n",
      "                                                                                                  \n",
      " dropout_9 (Dropout)            (None, 9, 64)        0           ['dense_12[0][0]']               \n",
      "                                                                                                  \n",
      " add_9 (Add)                    (None, 9, 64)        0           ['dropout_9[0][0]',              \n",
      "                                                                  'add_8[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_13 (LayerN  (None, 9, 64)       128         ['add_9[0][0]']                  \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " multi_head_attention_5 (MultiH  (None, 9, 64)       66368       ['layer_normalization_13[0][0]', \n",
      " eadAttention)                                                    'layer_normalization_13[0][0]'] \n",
      "                                                                                                  \n",
      " add_10 (Add)                   (None, 9, 64)        0           ['multi_head_attention_5[0][0]', \n",
      "                                                                  'add_9[0][0]']                  \n",
      "                                                                                                  \n",
      " layer_normalization_14 (LayerN  (None, 9, 64)       128         ['add_10[0][0]']                 \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " dense_13 (Dense)               (None, 9, 128)       8320        ['layer_normalization_14[0][0]'] \n",
      "                                                                                                  \n",
      " dropout_10 (Dropout)           (None, 9, 128)       0           ['dense_13[0][0]']               \n",
      "                                                                                                  \n",
      " dense_14 (Dense)               (None, 9, 64)        8256        ['dropout_10[0][0]']             \n",
      "                                                                                                  \n",
      " dropout_11 (Dropout)           (None, 9, 64)        0           ['dense_14[0][0]']               \n",
      "                                                                                                  \n",
      " add_11 (Add)                   (None, 9, 64)        0           ['dropout_11[0][0]',             \n",
      "                                                                  'add_10[0][0]']                 \n",
      "                                                                                                  \n",
      " layer_normalization_15 (LayerN  (None, 9, 64)       128         ['add_11[0][0]']                 \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " multi_head_attention_6 (MultiH  (None, 9, 64)       66368       ['layer_normalization_15[0][0]', \n",
      " eadAttention)                                                    'layer_normalization_15[0][0]'] \n",
      "                                                                                                  \n",
      " add_12 (Add)                   (None, 9, 64)        0           ['multi_head_attention_6[0][0]', \n",
      "                                                                  'add_11[0][0]']                 \n",
      "                                                                                                  \n",
      " layer_normalization_16 (LayerN  (None, 9, 64)       128         ['add_12[0][0]']                 \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " dense_15 (Dense)               (None, 9, 128)       8320        ['layer_normalization_16[0][0]'] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " dropout_12 (Dropout)           (None, 9, 128)       0           ['dense_15[0][0]']               \n",
      "                                                                                                  \n",
      " dense_16 (Dense)               (None, 9, 64)        8256        ['dropout_12[0][0]']             \n",
      "                                                                                                  \n",
      " dropout_13 (Dropout)           (None, 9, 64)        0           ['dense_16[0][0]']               \n",
      "                                                                                                  \n",
      " add_13 (Add)                   (None, 9, 64)        0           ['dropout_13[0][0]',             \n",
      "                                                                  'add_12[0][0]']                 \n",
      "                                                                                                  \n",
      " layer_normalization_17 (LayerN  (None, 9, 64)       128         ['add_13[0][0]']                 \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " multi_head_attention_7 (MultiH  (None, 9, 64)       66368       ['layer_normalization_17[0][0]', \n",
      " eadAttention)                                                    'layer_normalization_17[0][0]'] \n",
      "                                                                                                  \n",
      " add_14 (Add)                   (None, 9, 64)        0           ['multi_head_attention_7[0][0]', \n",
      "                                                                  'add_13[0][0]']                 \n",
      "                                                                                                  \n",
      " layer_normalization_18 (LayerN  (None, 9, 64)       128         ['add_14[0][0]']                 \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " dense_17 (Dense)               (None, 9, 128)       8320        ['layer_normalization_18[0][0]'] \n",
      "                                                                                                  \n",
      " dropout_14 (Dropout)           (None, 9, 128)       0           ['dense_17[0][0]']               \n",
      "                                                                                                  \n",
      " dense_18 (Dense)               (None, 9, 64)        8256        ['dropout_14[0][0]']             \n",
      "                                                                                                  \n",
      " dropout_15 (Dropout)           (None, 9, 64)        0           ['dense_18[0][0]']               \n",
      "                                                                                                  \n",
      " add_15 (Add)                   (None, 9, 64)        0           ['dropout_15[0][0]',             \n",
      "                                                                  'add_14[0][0]']                 \n",
      "                                                                                                  \n",
      " layer_normalization_19 (LayerN  (None, 9, 64)       128         ['add_15[0][0]']                 \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " flatten (Flatten)              (None, 576)          0           ['layer_normalization_19[0][0]'] \n",
      "                                                                                                  \n",
      " dropout_16 (Dropout)           (None, 576)          0           ['flatten[0][0]']                \n",
      "                                                                                                  \n",
      " dense_19 (Dense)               (None, 2048)         1181696     ['dropout_16[0][0]']             \n",
      "                                                                                                  \n",
      " dropout_17 (Dropout)           (None, 2048)         0           ['dense_19[0][0]']               \n",
      "                                                                                                  \n",
      " dense_20 (Dense)               (None, 1024)         2098176     ['dropout_17[0][0]']             \n",
      "                                                                                                  \n",
      " dropout_18 (Dropout)           (None, 1024)         0           ['dense_20[0][0]']               \n",
      "                                                                                                  \n",
      " dense_21 (Dense)               (None, 2)            2050        ['dropout_18[0][0]']             \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 4,734,729\n",
      "Trainable params: 4,734,722\n",
      "Non-trainable params: 7\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "new_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "198f1c86",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59/59 - 7s - loss: 1.3394 - accuracy: 0.6548 - 7s/epoch - 122ms/step\n",
      "Restored model, accuracy: 65.48%\n",
      "59/59 [==============================] - 7s 89ms/step\n",
      "(1857, 2)\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the restored model\n",
    "loss, acc = new_model.evaluate(test_x, test_y, verbose=2)\n",
    "print('Restored model, accuracy: {:5.2f}%'.format(100 * acc))\n",
    "\n",
    "print(new_model.predict(test_x).shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f183b126",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['assets', 'keras_metadata.pb', 'saved_model.pb', 'variables']"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "os.listdir(\"C:/Users/GB/Desktop/dataset/model\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e7f5aa21",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59/59 - 7s - loss: 1.3394 - accuracy: 0.6548 - 7s/epoch - 117ms/step\n",
      "Restored model, accuracy: 65.48%\n"
     ]
    }
   ],
   "source": [
    "# Save the weights\n",
    "vit.save_weights('./checkpoints/my_checkpoint')\n",
    "\n",
    "\n",
    "\n",
    "# Restore the weights\n",
    "vit.load_weights('./checkpoints/my_checkpoint')\n",
    "\n",
    "# Evaluate the model\n",
    "loss, acc = vit.evaluate(test_x, test_y, verbose=2)\n",
    "print(\"Restored model, accuracy: {:5.2f}%\".format(100 * acc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "00018c10",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'vit' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[2], line 2\u001b[0m\n\u001b[0;32m      1\u001b[0m get_ipython()\u001b[38;5;241m.\u001b[39msystem(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmkdir -p saved_model\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m----> 2\u001b[0m vit\u001b[38;5;241m.\u001b[39msave(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124msaved_model/this_model\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'vit' is not defined"
     ]
    }
   ],
   "source": [
    "!mkdir -p saved_model\n",
    "vit.save('saved_model/this_model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7090ec56",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'vit' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Input \u001b[1;32mIn [2]\u001b[0m, in \u001b[0;36m<cell line: 3>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mcv2\u001b[39;00m\n\u001b[1;32m----> 3\u001b[0m prediction \u001b[38;5;241m=\u001b[39m \u001b[43mvit\u001b[49m\u001b[38;5;241m.\u001b[39mpredict({\u001b[38;5;124m'\u001b[39m\u001b[38;5;124minput\u001b[39m\u001b[38;5;124m'\u001b[39m: test_image })\n",
      "\u001b[1;31mNameError\u001b[0m: name 'vit' is not defined"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "\n",
    "prediction = vit.predict({'input': test_image })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "537c9abd",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}