Code/XAI_TL_gradcam.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8870f399",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import load_model\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.utils import load_img, img_to_array, array_to_img\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.applications.vgg19 import preprocess_input, decode_predictions\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import numpy as np\n",
    "from IPython.display import Image, display\n",
    "import matplotlib.cm as cm\n",
    "\n",
    "#http://gradcam.cloudcv.org/\n",
    "#https://keras.io/examples/vision/grad_cam/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7df286b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_img_array(img_path, size):\n",
    "    # `img` is a PIL image of size 299x299\n",
    "    img = load_img(img_path, target_size=size)\n",
    "    # `array` is a float32 Numpy array of shape (299, 299, 3)\n",
    "    array = img_to_array(img)\n",
    "    # We add a dimension to transform our array into a \"batch\"\n",
    "    # of size (1, 299, 299, 3)\n",
    "    array = np.expand_dims(array, axis=0)\n",
    "    return array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cce7dd5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_gradcam_heatmap(img_array, model, last_conv_layer_name, pred_index=None):\n",
    "    # First, we create a model that maps the input image to the activations\n",
    "    # of the last conv layer as well as the output predictions\n",
    "    grad_model = tf.keras.models.Model(\n",
    "        [model.inputs], [model.get_layer(last_conv_layer_name).output, model.output]\n",
    "    )\n",
    "\n",
    "    # Then, we compute the gradient of the top predicted class for our input image\n",
    "    # with respect to the activations of the last conv layer\n",
    "    with tf.GradientTape() as tape:\n",
    "        last_conv_layer_output, preds = grad_model(img_array)\n",
    "        if pred_index is None:\n",
    "            pred_index = tf.argmax(preds[0])\n",
    "        class_channel = preds[:, pred_index]\n",
    "\n",
    "    # This is the gradient of the output neuron (top predicted or chosen)\n",
    "    # with regard to the output feature map of the last conv layer\n",
    "    grads = tape.gradient(class_channel, last_conv_layer_output)\n",
    "\n",
    "    # This is a vector where each entry is the mean intensity of the gradient\n",
    "    # over a specific feature map channel\n",
    "    pooled_grads = tf.reduce_mean(grads, axis=(0, 1, 2))\n",
    "\n",
    "    # We multiply each channel in the feature map array\n",
    "    # by \"how important this channel is\" with regard to the top predicted class\n",
    "    # then sum all the channels to obtain the heatmap class activation\n",
    "    last_conv_layer_output = last_conv_layer_output[0]\n",
    "    heatmap = last_conv_layer_output @ pooled_grads[..., tf.newaxis]\n",
    "    heatmap = tf.squeeze(heatmap)\n",
    "\n",
    "    # For visualization purpose, we will also normalize the heatmap between 0 & 1\n",
    "    heatmap = tf.maximum(heatmap, 0) / tf.math.reduce_max(heatmap)\n",
    "    \n",
    "    return heatmap.numpy()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20bdb974",
   "metadata": {},
   "source": [
    "### Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9a2915f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Leemos el modelo guardado\n",
    "target_size = (224, 224)\n",
    "path_image = '../../dataset/classification/PCB_Dataset_Split/test/ig/'\n",
    "name_image = 'MMY_790179.jpg'\n",
    "\n",
    "model = load_model('../../model/classification/vgg19_trainable_true_best_model.h5')\n",
    "last_conv_layer_name = \"block5_conv4\"\n",
    "decode_predictions = decode_predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb993c18",
   "metadata": {},
   "source": [
    "### Carga y ploteo de la imagen de entrada"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "40bb9ce4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-0.5, 223.5, 223.5, -0.5)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#load the image\n",
    "my_image = load_img(path_image+name_image, target_size = (224, 224))\n",
    "plt.imshow(my_image)\n",
    "plt.axis('off')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88a3c1d6",
   "metadata": {},
   "source": [
    "### Gradcam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "66fbd784",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 5s 5s/step\n",
      "Predicted: [[ 105.83373 1942.6595  -684.7925  -260.58884 -655.6293  -212.28801\n",
      "   665.1183  -965.6527 ]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Prepare image\n",
    "img_array = preprocess_input(get_img_array(path_image+name_image, size = target_size))\n",
    "\n",
    "# Make model\n",
    "#model = model_builder(weights=\"imagenet\")\n",
    "\n",
    "# Remove last layer's softmax\n",
    "model.layers[-1].activation = None\n",
    "\n",
    "# Print what the top predicted class is\n",
    "preds = model.predict(img_array)\n",
    "print(\"Predicted:\", preds)\n",
    "\n",
    "# Generate class activation heatmap\n",
    "heatmap = make_gradcam_heatmap(img_array, model, last_conv_layer_name)\n",
    "\n",
    "# Display heatmap\n",
    "plt.matshow(heatmap)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94aa2cee",
   "metadata": {},
   "source": [
    "### Utilización de la salida de gradcam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "108eb853",
   "metadata": {},
   "outputs": [],
   "source": [
    "def save_and_display_gradcam(img_path, heatmap, cam_path = \"../../images/XAI/\"+name_image, alpha = 0.4):\n",
    "    # Load the original image\n",
    "    img = load_img(img_path)\n",
    "    img = img_to_array(img)\n",
    "\n",
    "    # Rescale heatmap to a range 0-255\n",
    "    heatmap = np.uint8(255 * heatmap)\n",
    "\n",
    "    # Use jet colormap to colorize heatmap\n",
    "    jet = cm.get_cmap(\"jet\")\n",
    "\n",
    "    # Use RGB values of the colormap\n",
    "    jet_colors = jet(np.arange(256))[:, :3]\n",
    "    jet_heatmap = jet_colors[heatmap]\n",
    "    print(jet_heatmap.shape)\n",
    "\n",
    "    # Create an image with RGB colorized heatmap\n",
    "    jet_heatmap = array_to_img(jet_heatmap)\n",
    "    jet_heatmap = jet_heatmap.resize((img.shape[1], img.shape[0]))\n",
    "    jet_heatmap = img_to_array(jet_heatmap)\n",
    "    \n",
    "    #rint(jet_heatmap)\n",
    "    print(jet_heatmap.shape)\n",
    "    print(img.shape)\n",
    "    \n",
    "    # Superimpose the heatmap on original image\n",
    "    superimposed_img = jet_heatmap * alpha + img\n",
    "    superimposed_img = array_to_img(superimposed_img)\n",
    "\n",
    "    # Save the superimposed image\n",
    "    superimposed_img.save(cam_path)\n",
    "\n",
    "    # Display Grad CAM\n",
    "    display(Image(cam_path))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "62c2a1ea",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(14, 14, 3)\n",
      "(363, 360, 3)\n",
      "(363, 360, 3)\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "save_and_display_gradcam(path_image+name_image, heatmap)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "644cd6ab",
   "metadata": {},
   "source": [
    "### Estudio del modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "fc9c801b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 224, 224, 3)]     0         \n",
      "                                                                 \n",
      " block1_conv1 (Conv2D)       (None, 224, 224, 64)      1792      \n",
      "                                                                 \n",
      " block1_conv2 (Conv2D)       (None, 224, 224, 64)      36928     \n",
      "                                                                 \n",
      " block1_pool (MaxPooling2D)  (None, 112, 112, 64)      0         \n",
      "                                                                 \n",
      " block2_conv1 (Conv2D)       (None, 112, 112, 128)     73856     \n",
      "                                                                 \n",
      " block2_conv2 (Conv2D)       (None, 112, 112, 128)     147584    \n",
      "                                                                 \n",
      " block2_pool (MaxPooling2D)  (None, 56, 56, 128)       0         \n",
      "                                                                 \n",
      " block3_conv1 (Conv2D)       (None, 56, 56, 256)       295168    \n",
      "                                                                 \n",
      " block3_conv2 (Conv2D)       (None, 56, 56, 256)       590080    \n",
      "                                                                 \n",
      " block3_conv3 (Conv2D)       (None, 56, 56, 256)       590080    \n",
      "                                                                 \n",
      " block3_conv4 (Conv2D)       (None, 56, 56, 256)       590080    \n",
      "                                                                 \n",
      " block3_pool (MaxPooling2D)  (None, 28, 28, 256)       0         \n",
      "                                                                 \n",
      " block4_conv1 (Conv2D)       (None, 28, 28, 512)       1180160   \n",
      "                                                                 \n",
      " block4_conv2 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
      "                                                                 \n",
      " block4_conv3 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
      "                                                                 \n",
      " block4_conv4 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
      "                                                                 \n",
      " block4_pool (MaxPooling2D)  (None, 14, 14, 512)       0         \n",
      "                                                                 \n",
      " block5_conv1 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
      "                                                                 \n",
      " block5_conv2 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
      "                                                                 \n",
      " block5_conv3 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
      "                                                                 \n",
      " block5_conv4 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
      "                                                                 \n",
      " block5_pool (MaxPooling2D)  (None, 7, 7, 512)         0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 25088)             0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 8)                 200712    \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 20,225,096\n",
      "Trainable params: 20,225,096\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "67c4dae1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['input_1',\n",
       " 'block1_conv1',\n",
       " 'block1_conv2',\n",
       " 'block1_pool',\n",
       " 'block2_conv1',\n",
       " 'block2_conv2',\n",
       " 'block2_pool',\n",
       " 'block3_conv1',\n",
       " 'block3_conv2',\n",
       " 'block3_conv3',\n",
       " 'block3_conv4',\n",
       " 'block3_pool',\n",
       " 'block4_conv1',\n",
       " 'block4_conv2',\n",
       " 'block4_conv3',\n",
       " 'block4_conv4',\n",
       " 'block4_pool',\n",
       " 'block5_conv1',\n",
       " 'block5_conv2',\n",
       " 'block5_conv3',\n",
       " 'block5_conv4',\n",
       " 'block5_pool',\n",
       " 'flatten',\n",
       " 'dense']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer_names = [layer.name for layer in model.layers]\n",
    "layer_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "235d663f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<keras.engine.input_layer.InputLayer at 0x26662e58c70>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662ea63a0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662ea65e0>,\n",
       " <keras.layers.pooling.max_pooling2d.MaxPooling2D at 0x26662ea6ac0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f37160>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f373a0>,\n",
       " <keras.layers.pooling.max_pooling2d.MaxPooling2D at 0x26662f378b0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f37f10>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f3a190>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f3a6a0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f3ac10>,\n",
       " <keras.layers.pooling.max_pooling2d.MaxPooling2D at 0x26662f3aee0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f3e820>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f3ea60>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f3ef70>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f3ef40>,\n",
       " <keras.layers.pooling.max_pooling2d.MaxPooling2D at 0x26662f41ac0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f44160>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f443a0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f448b0>,\n",
       " <keras.layers.convolutional.conv2d.Conv2D at 0x26662f44e20>,\n",
       " <keras.layers.pooling.max_pooling2d.MaxPooling2D at 0x26662f44fa0>,\n",
       " <keras.layers.reshaping.flatten.Flatten at 0x26662f38c10>,\n",
       " <keras.layers.core.dense.Dense at 0x26662f38d90>]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "2992824b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<KerasTensor: shape=(None, 224, 224, 3) dtype=float32 (created by layer 'input_1')>,\n",
       " <KerasTensor: shape=(None, 224, 224, 64) dtype=float32 (created by layer 'block1_conv1')>,\n",
       " <KerasTensor: shape=(None, 224, 224, 64) dtype=float32 (created by layer 'block1_conv2')>,\n",
       " <KerasTensor: shape=(None, 112, 112, 64) dtype=float32 (created by layer 'block1_pool')>,\n",
       " <KerasTensor: shape=(None, 112, 112, 128) dtype=float32 (created by layer 'block2_conv1')>,\n",
       " <KerasTensor: shape=(None, 112, 112, 128) dtype=float32 (created by layer 'block2_conv2')>,\n",
       " <KerasTensor: shape=(None, 56, 56, 128) dtype=float32 (created by layer 'block2_pool')>,\n",
       " <KerasTensor: shape=(None, 56, 56, 256) dtype=float32 (created by layer 'block3_conv1')>,\n",
       " <KerasTensor: shape=(None, 56, 56, 256) dtype=float32 (created by layer 'block3_conv2')>,\n",
       " <KerasTensor: shape=(None, 56, 56, 256) dtype=float32 (created by layer 'block3_conv3')>,\n",
       " <KerasTensor: shape=(None, 56, 56, 256) dtype=float32 (created by layer 'block3_conv4')>,\n",
       " <KerasTensor: shape=(None, 28, 28, 256) dtype=float32 (created by layer 'block3_pool')>,\n",
       " <KerasTensor: shape=(None, 28, 28, 512) dtype=float32 (created by layer 'block4_conv1')>,\n",
       " <KerasTensor: shape=(None, 28, 28, 512) dtype=float32 (created by layer 'block4_conv2')>,\n",
       " <KerasTensor: shape=(None, 28, 28, 512) dtype=float32 (created by layer 'block4_conv3')>,\n",
       " <KerasTensor: shape=(None, 28, 28, 512) dtype=float32 (created by layer 'block4_conv4')>,\n",
       " <KerasTensor: shape=(None, 14, 14, 512) dtype=float32 (created by layer 'block4_pool')>,\n",
       " <KerasTensor: shape=(None, 14, 14, 512) dtype=float32 (created by layer 'block5_conv1')>,\n",
       " <KerasTensor: shape=(None, 14, 14, 512) dtype=float32 (created by layer 'block5_conv2')>,\n",
       " <KerasTensor: shape=(None, 14, 14, 512) dtype=float32 (created by layer 'block5_conv3')>,\n",
       " <KerasTensor: shape=(None, 14, 14, 512) dtype=float32 (created by layer 'block5_conv4')>,\n",
       " <KerasTensor: shape=(None, 7, 7, 512) dtype=float32 (created by layer 'block5_pool')>,\n",
       " <KerasTensor: shape=(None, 25088) dtype=float32 (created by layer 'flatten')>,\n",
       " <KerasTensor: shape=(None, 8) dtype=float32 (created by layer 'dense')>]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer_outputs = [layer.output for layer in model.layers]\n",
    "layer_outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "70d31fb0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<keras.engine.functional.Functional at 0x266518e6ee0>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Feature Map\n",
    "feature_map_model = tf.keras.models.Model(inputs = model.input, outputs = layer_outputs)\n",
    "feature_map_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "940ef775",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 212ms/step\n"
     ]
    }
   ],
   "source": [
    "feature_maps = feature_map_model.predict(my_image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "46846d05",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The shape of the input_1 is =======>> (1, 224, 224, 3)\n",
      "The shape of the block1_conv1 is =======>> (1, 224, 224, 64)\n",
      "The shape of the block1_conv2 is =======>> (1, 224, 224, 64)\n",
      "The shape of the block1_pool is =======>> (1, 112, 112, 64)\n",
      "The shape of the block2_conv1 is =======>> (1, 112, 112, 128)\n",
      "The shape of the block2_conv2 is =======>> (1, 112, 112, 128)\n",
      "The shape of the block2_pool is =======>> (1, 56, 56, 128)\n",
      "The shape of the block3_conv1 is =======>> (1, 56, 56, 256)\n",
      "The shape of the block3_conv2 is =======>> (1, 56, 56, 256)\n",
      "The shape of the block3_conv3 is =======>> (1, 56, 56, 256)\n",
      "The shape of the block3_conv4 is =======>> (1, 56, 56, 256)\n",
      "The shape of the block3_pool is =======>> (1, 28, 28, 256)\n",
      "The shape of the block4_conv1 is =======>> (1, 28, 28, 512)\n",
      "The shape of the block4_conv2 is =======>> (1, 28, 28, 512)\n",
      "The shape of the block4_conv3 is =======>> (1, 28, 28, 512)\n",
      "The shape of the block4_conv4 is =======>> (1, 28, 28, 512)\n",
      "The shape of the block4_pool is =======>> (1, 14, 14, 512)\n",
      "The shape of the block5_conv1 is =======>> (1, 14, 14, 512)\n",
      "The shape of the block5_conv2 is =======>> (1, 14, 14, 512)\n",
      "The shape of the block5_conv3 is =======>> (1, 14, 14, 512)\n",
      "The shape of the block5_conv4 is =======>> (1, 14, 14, 512)\n",
      "The shape of the block5_pool is =======>> (1, 7, 7, 512)\n",
      "The shape of the flatten is =======>> (1, 25088)\n",
      "The shape of the dense is =======>> (1, 8)\n"
     ]
    }
   ],
   "source": [
    "for layer_name, feature_map in zip(layer_names, feature_maps):\n",
    "    print(f\"The shape of the {layer_name} is =======>> {feature_map.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a3261d02",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}