Create GRADCAM.py

Author: ashishpatel26

GRADCAM.py

@@ -0,0 +1,90 @@
+'''
+Install Grad CAM : `!pip install tf-explain`
+* src : https://github.com/sicara/tf-explain
+* paper : Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
+* Reference : https://arxiv.org/abs/1610.02391
+* Abstract : We propose a technique for producing "visual explanations" for decisions from a large class
+  of CNN-based models, making them more transparent. Our approach - Gradient-weighted Class Activation Mapping 
+  (Grad-CAM), uses the gradients of any target concept, flowing into the final convolutional layer to produce 
+  a coarse localization map highlighting important regions in the image for predicting the concept. Grad-CAM 
+  is applicable to a wide variety of CNN model-families: 
+  (1) CNNs with fully-connected layers, 
+  (2) CNNs used for structured outputs, 
+  (3) CNNs used in tasks with multimodal inputs or reinforcement learning, 
+  without any architectural changes or re-training. We combine Grad-CAM with fine-grained visualizations to create
+   a high-resolution class-discriminative visualization and apply it to off-the-shelf image classification, captioning, 
+   and visual question answering (VQA) models, including ResNet-based architectures. In the context of image classification 
+   models, our visualizations (a) lend insights into their failure modes, 
+   (b) are robust to adversarial images, (c) outperform previous methods on localization, (d) are more faithful to the 
+   underlying model and (e) help achieve generalization by identifying dataset bias. For captioning and VQA, we show that even
+    non-attention based models can localize inputs. We devise a way to identify important neurons through Grad-CAM and combine it 
+    with neuron names to provide textual explanations for model decisions. Finally, we design and conduct human studies to measure 
+    if Grad-CAM helps users establish appropriate trust in predictions from models and show that Grad-CAM helps untrained users 
+    successfully discern a 'stronger' nodel from a 'weaker' one even when both make identical predictions. 
+
+##### Note : you can pass `model` object as any tensorflow keras model.
+'''
+import tensorflow as tf
+import os
+import cv2
+from tensorflow.keras.preprocessing.image import *
+from tensorflow.keras.applications.imagenet_utils import preprocess_input
+import matplotlib.pyplot as plt
+from tf_explain.core.grad_cam import GradCAM
+import numpy as np
+
+
+class_mapping = {0: 'Cat', 1: 'Dog'}
+
+
+def preprocessing_image(instancePath):
+    original_image = plt.imread(instancePath)
+    image = load_img(instancePath, target_size=(224, 224))
+    image = img_to_array(image)
+    image = tf.expand_dims(image, 0)
+    image /= 255.0
+    image = preprocess_input(image)
+    return image, original_image
+
+
+def predict_per(IMAGE_PATH):
+    image, o_image = preprocessing_image(IMAGE_PATH)
+    prediction = np.argmax(model.predict(image))
+    prediction_per = np.max(model.predict(image))
+    return class_mapping[prediction], prediction, prediction_per
+
+
+def Grad_cam_vis(IMAGE_PATH, OUTPUT_PATH, ACTUAL_LABEL: str):
+    prediction, index, confidense = predict_per(IMAGE_PATH)
+    img = tf.keras.preprocessing.image.load_img(IMAGE_PATH, target_size=(224, 224))
+    img = tf.keras.preprocessing.image.img_to_array(img)
+    data = ([img], None)
+
+    # Start explainer
+    explainer = GradCAM()
+    grid = explainer.explain(data, model, class_index=index)
+
+    explainer.save(grid, ".", OUTPUT_PATH)
+
+    im = cv2.cvtColor(cv2.imread(IMAGE_PATH), cv2.COLOR_BGR2RGB)
+    im1 = cv2.cvtColor(cv2.imread(OUTPUT_PATH), cv2.COLOR_BGR2RGB)
+    plt.figure(figsize=(15, 8))
+    plt.subplot(1, 2, 1)
+    plt.imshow(im)
+    plt.xlabel(f"Actaul: Healthy")
+    plt.subplot(1, 2, 2)
+    plt.imshow(im1)
+    plt.xlabel(f"predict:{prediction}\nConfidence: {confidense}")
+    plt.show()
+
+
+images = os.listdir('/content/Images/')
+for img in images:
+  # print(img)
+  os.mkdirs("Output", exist_ok = True)
+  maping_result  = {'c':'Cat', 'd':'Dog'}
+  actual_label = maping_result[img[0].lower()]
+  IMAGE_PATH = f"/content/Images/{img}"
+  OUTPUT_PATH = f"/content/Output/output_{img[:-4]}.jpg"
+  pred_class, prediction, prediction_per = predict_per(IMAGE_PATH)
+  Grad_cam_vis(IMAGE_PATH, OUTPUT_PATH, actual_label)