Tidy

app.py

@@ -65,7 +65,7 @@ def fn(video: gr.Video, actions: list[int]):
             time.sleep(2)
             toggle = False
       if i % config.classify_action_frame_step != 0 and i % config.detect_object_frame_step != 0:
-          time.sleep(0.3)
+          time.sleep(0.2)
     processed_clip = ImageSequenceClip(processed_frames, clip.fps)
     processed_clip.audio = clip.audio
     processed_clip.write_videofile(output, fps=clip.fps, audio_codec='aac', logger=None)

inference.py

@@ -72,7 +72,7 @@ def draw_classes(frame, actions):
     label=labels,
     image=frame,
     label_names=label_names,
-    font_size=10,
+    font_size=20,
     alpha=0)
   return frame