update app.py

app.py

@@ -51,14 +51,15 @@ def predict(input_file, downsample_size):
                 in_img = torch.from_numpy(in_img).float()
                 encoded = avg_pool(model(in_img))[0, :, 0, 0].cpu().numpy()
                 img_vecs += [encoded]
-
     img_vecs = np.asarray(img_vecs)
+    print("images encoded")
     rv_indices, _ = furthest_neighbours(
         img_vecs,
         downsample_size,
         seed=0)
     indices = np.zeros((img_vecs.shape[0],))
     indices[np.asarray(rv_indices)] = 1
+    print("images selected")
 
     global_ctr = 0
     for fp_i, file_path in enumerate([input_file]):
@@ -69,6 +70,7 @@ def predict(input_file, downsample_size):
             if indices[global_ctr] == 1:
                 cv2.imwrite(join(selected_directory, str(global_ctr) + ".jpg"), img)
             global_ctr += 1
+    print("selected images extracted")
 
     all_selected_imgs_path = [join(selected_directory, f) for f in os.listdir(selected_directory) if
                               isfile(join(selected_directory, f))]
@@ -77,6 +79,7 @@ def predict(input_file, downsample_size):
     for i, f in enumerate(all_selected_imgs_path):
         zipf.write(f, basename(f))
     zipf.close()
+    print("selected images zipped")
 
     return zip_path
 
@@ -87,4 +90,4 @@ demo = gr.Interface(
             gr.inputs.Number(label="Downsample size")],
     outputs=gr.outputs.File(label="Zip"))
 
-demo.launch()
+demo.launch(cache_examples=False)