revert -6

yolov5.py

@@ -90,7 +90,6 @@ def xai_yolov5(image,target_lyr = -5, n_components = 8):
 
     rgb_img_float, batch_explanations, result = dff_nmf(image, target_lyr = -5, n_components = 8)
     #result = np.hstack(result)
-    print("sample...",rgb_img_float.shape,batch_explanations.shape )
     im = visualize_batch_explanations(rgb_img_float, batch_explanations)  ##########to be displayed
     
     # Combine results
@@ -234,21 +233,23 @@ def dff_nmf(image, target_lyr, n_components):
 
         fig.canvas.draw()  # Draw the canvas to make sure the image is rendered
         image_array = np.array(fig.canvas.renderer.buffer_rgba())  # Convert to numpy array
+        print("____________image_arrya", image_array.shape)
         image_resized = cv2.resize(image_array, (640, 640))
         rgba_channels = cv2.split(image_resized)
-        alpha_channel = rgba_channels[3] 
+        alpha_channel = rgba_channels[3] / 255.0
         rgb_channels = np.stack(rgba_channels[:3], axis=-1)
         overlay_img = (alpha_channel[..., None] * image) + ((1 - alpha_channel[..., None]) * rgb_channels)
-        print("shape....", rgb_img_float.shape, rgb_channels.shape)
-        #visualization = show_factorization_on_image(rgb_img_float, rgb_channels , image_weight=0.3)
         
         #temp = image_array.reshape((rgb_img_float.shape[0],rgb_img_float.shape[1]) )
         #visualization = show_factorization_on_image(rgb_img_float, image_array.resize((rgb_img_float.shape)) , image_weight=0.3)
         results.append(overlay_img)
         plt.clf()  
         #return image_array
+
+
     return rgb_img_float, batch_explanations, results
 
+
 def visualize_batch_explanations(rgb_img_float, batch_explanations, image_weight=0.7):
     for i, explanation in enumerate(batch_explanations):
         # Create visualization for each explanation
@@ -260,4 +261,4 @@ def visualize_batch_explanations(rgb_img_float, batch_explanations, image_weight
         plt.show()  # Show the plot
     plt.savefig("test_w.png")
     print('viz', visualization.shape)
-    return visualization
+    return visualization