Update app.py

app.py

@@ -17,10 +17,11 @@ base_model.trainable = False
 
 
 def img_pros(img):
-  img = tf.keras.preprocessing.image.img_to_array(img)
-  img = tf.image.resize(img, [160,160])
+  original_img = tf.keras.preprocessing.image.img_to_array(img)
+  original_img_shape = img.shape
+  img = tf.image.resize(original_img, [160,160])
   img = tf.expand_dims(img, axis = 0)
-  return img
+  return img, original_img, original_img_shape
 
 #function for creating model
 #returns model, its inputs, Xception's last conv output, the whole model's outputs
@@ -102,7 +103,7 @@ def superimpose_single(heatmap, img, alpha = 0.7):
     
     # Create an image with RGB colorized heatmap
     jet_heatmap = keras.utils.array_to_img(jet_heatmap)
-    jet_heatmap = jet_heatmap.resize((160,160))
+    jet_heatmap = jet_heatmap.resize((original_img_shape[1], original_img_shape[0]))
     jet_heatmap = keras.utils.img_to_array(jet_heatmap)
     
     # Superimpose the heatmap on original image
@@ -112,20 +113,20 @@ def superimpose_single(heatmap, img, alpha = 0.7):
     return superimposed_img
 
 #for generating single gradcam image
-def gen_grad_img_single(grad_model, img, class_index, alpha = 0.4):
+def gen_grad_img_single(grad_model, img, class_index, original_img, original_img_shape, alpha = 0.4):
     heatmaps, y_pred = create_heatmap(grad_model, img, class_index)
     
     # for i in range(len(y_pred)):
     #     if y_pred[i] > 0.5: y_pred[i] = 1
     #     else: y_pred[i] = 0
     
-    img = superimpose_single(heatmaps, img[0])
+    img = superimpose_single(heatmaps, original_img, original_img_shape)
     return np.array(img).astype('uint8'), y_pred
 
 
-def gen_grad_both(grad_model, img):
-    img_c, y_pred_c = gen_grad_img_single(grad_model, img, 0)
-    img_d, y_pred_d = gen_grad_img_single(grad_model, img, 1)
+def gen_grad_both(grad_model, img, original_img, original_img_shape):
+    img_c, y_pred_c = gen_grad_img_single(grad_model, img, 0, original_img, original_img_shape)
+    img_d, y_pred_d = gen_grad_img_single(grad_model, img, 1, original_img, original_img_shape)
     y_pred_c = np.around(y_pred_c,3)
     y_pred_d = np.around(y_pred_d,3)
     # show_imgs([img_c, img_d], [y_true, y_true], [size[0], size[1]], cols, [y_pred_c, y_pred_d], font_size = font_size)
@@ -139,9 +140,9 @@ def gen_grad_both(grad_model, img):
 weights = "weights_nm.h5"
 
 def get_grad(img):
-    img = img_pros(img)
+    img, original_img, original_img_shape = img_pros(img)
     grad_model = create_grad_model(weights, 2, "softmax")
-    grad_img_c, grad_img_d, y_pred, infer = gen_grad_both(grad_model, img)
+    grad_img_c, grad_img_d, y_pred, infer = gen_grad_both(grad_model, img, original_img, original_img_shape)
     # pred_class = ""
     # if y_pred[0] > 0.5: pred_class = "cat"
     # else: pred_class = "dog"