update app

app.py

@@ -49,7 +49,7 @@ def image_classifier(input_image, top_classes=3, show_cam=True, target_layer=-2,
     targets = [ClassifierOutputTarget(label)]
     grayscale_cam = grad_cam(input_tensor=input, targets=targets)
     grayscale_cam = grayscale_cam[0, :]
-    output_image = show_cam_on_image(input_image/255, grayscale_cam, use_rgb=True, image_weight=transparency)
+    output_image = show_cam_on_image(input_image / 255, grayscale_cam, use_rgb=True, image_weight=transparency)
 
     return output_image if show_cam else input_image, confidences
 
@@ -58,10 +58,10 @@ demo1 = gr.Interface(
     fn=image_classifier,
     inputs=[
         gr.Image(shape=(32, 32), label="Input Image", value='examples/cat.jpg'),
-        gr.Slider(1, 10, value = 3, step=1, label="Number of Top Classes"),
+        gr.Slider(1, 10, value=3, step=1, label="Number of Top Classes"),
         gr.Checkbox(label="Show GradCAM?", value=True),
-        gr.Slider(-4, -1, value = -2, step=1, label="Which Layer?"),
-        gr.Slider(0, 1, value = 0.7, label="Transparency", step=0.1)
+        gr.Slider(-4, -1, value=-2, step=1, label="Which Layer?"),
+        gr.Slider(0, 1, value=0.7, label="Transparency", step=0.1)
     ],
     outputs=[gr.Image(shape=(32, 32), label="Output Image"),
              gr.Label(label='Top Classes')],
@@ -69,13 +69,18 @@ demo1 = gr.Interface(
 )
 
 
-def show_incorrect(num_examples=10, show_cam=True, target_layer=-1, transparency=0.5):
+def show_incorrect(num_examples=20, show_cam=True, target_layer=-1, transparency=0.5):
     result = list()
-    for index, row in missed_df.head(num_examples).iterrows():
+    for index, row in missed_df.iterrows():
         image = np.asarray(Image.open(f'missed_examples/{index}.jpg'))
-        output_image, confidence = image_classifier(image, show_cam=show_cam, target_layer=target_layer, transparency=transparency)
+        output_image, confidence = image_classifier(image, show_cam=show_cam, target_layer=target_layer,
+                                                    transparency=transparency)
+        truth = row['ground_truths']
         predicted = list(confidence)[0]
-        result.append((output_image, f"{row['ground_truths']} / {row['predicted_vals']}"))
+        if truth != predicted:
+            result.append((output_image, f"{row['ground_truths']} / {predicted}"))
+        if len(result) >= num_examples:
+            break
     return result
 
 
@@ -84,8 +89,8 @@ demo2 = gr.Interface(
     inputs=[
         gr.Number(value=20, minimum=1, maximum=len(missed_df), label="No. of missclassified Examples", precision=0),
         gr.Checkbox(label="Show GradCAM?", value=True),
-        gr.Slider(-4, -1, value = -2, step=1, label="Which Layer?"),
-        gr.Slider(0, 1, value = 0.7, label="Transparency", step=0.1),
+        gr.Slider(-4, -1, value=-2, step=1, label="Which Layer?"),
+        gr.Slider(0, 1, value=0.7, label="Transparency", step=0.1),
     ],
     outputs=[gr.Gallery(label="Missclassified Images (Truth / Predicted)", columns=4)]
 )