Update app.py

app.py

@@ -1,17 +1,11 @@
 import warnings
 warnings.filterwarnings("ignore")
 import gradio as gr
-from src.video_model import describe_video  # Your video processing function
-from src.text_processor import process_description  # Your text processing function
+from src.video_model import describe_video  # Assuming this function processes the video and query
 
-# --- Global variable to store the prediction ---
-prediction = None
-
-# --- Function to handle video processing ---
-def process_video(video, sitting, hands, location, screen):
-    global prediction  # Access the global prediction variable
-
-    query = "Describe this video in detail and answer the questions."
+# --- Function to construct the final query --- 
+def process_video_and_questions(video, sitting, hands, location, screen):
+    query = "Describe this video in detail and answer the questions"
     additional_info = []
     if sitting:
         additional_info.append("Is the subject in the video standing or sitting?")
@@ -21,30 +15,23 @@ def process_video(video, sitting, hands, location, screen):
         additional_info.append("Is the subject present indoors or outdoors?")
     if screen:
         additional_info.append("Is the subject interacting with a screen in the background by facing the screen?")
-
+    
     final_query = query + " " + " ".join(additional_info)
-    prediction = describe_video(video, final_query)
-    # Enable the "Process Text" button
-    return gr.update(visible=True), prediction 
+    # Assuming your describe_video function handles the video processing
+    response = describe_video(video, final_query) 
+    return response
 
-# --- Function to trigger text processing ---
-def process_and_display_text():
-    global prediction 
-    json_response = process_description(prediction)
-    return json_response
+# Video and text inputs for the interface
+video = gr.Video(label="Video")
 
-# ... (Gradio interface code) ... 
+# Options as checkboxes
+sitting = gr.Checkbox(label="Sitting/Standing")
+hands = gr.Checkbox(label="Hands Free/Not Free")
+location = gr.Checkbox(label="Indoors/Outdoors")
+screen = gr.Checkbox(label="Screen Interaction")
 
-#video = gr.Video(label="Video")
-#sitting = gr.Checkbox(label="Sitting/Standing")
-#hands = gr.Checkbox(label="Hands Free/Not Free")
-#location = gr.Checkbox(label="Indoors/Outdoors")
-#screen = gr.Checkbox(label="Screen Interaction")
-
-# Output components
-#video_description = gr.Textbox(label="Video Description")
-#json_output = gr.JSON(label="JSON Output")
-#process_button = gr.Button("Process Text", visible=False) 
+# Output for the interface
+response = gr.Textbox(label="Predicted answer", show_label=True, show_copy_button=True)
 
 # Examples for the interface
 examples = [
@@ -66,39 +53,22 @@ title = "GSoC Super Raid Annotator"
 description = "Annotate Videos"
 article = "<p style='text-align: center'><a href='https://github.com/OpenBMB/MiniCPM-V' target='_blank'>Model GitHub Repo</a> | <a href='https://huggingface.co/openbmb/MiniCPM-V-2_6' target='_blank'>Model Page</a></p>"
 
+
 custom_theme = gr.themes.Soft(
+    # Set the primary hue of the Soft theme to your red color
     primary_hue="red", 
-    secondary_hue="red"
+    secondary_hue="red")
+
+# Launch the interface
+interface = gr.Interface(
+    fn=process_video_and_questions, # Updated function to handle the query construction
+    inputs=[video, sitting, hands, location, screen], 
+    outputs=response,
+    examples=examples,
+    title=title,
+    description=description,
+    article=article,
+    theme=custom_theme,
+    allow_flagging="never",
 )
-
-
-with gr.Blocks(theme=custom_theme) as demo:  # Use 'with' for gr.Blocks
-    video = gr.Video(label="Video")
-    sitting = gr.Checkbox(label="Sitting/Standing")
-    hands = gr.Checkbox(label="Hands Free/Not Free")
-    location = gr.Checkbox(label="Indoors/Outdoors")
-    screen = gr.Checkbox(label="Screen Interaction")
-
-    # Output components
-    video_description = gr.Textbox(label="Video Description")
-    json_output = gr.JSON(label="JSON Output")
-    process_button = gr.Button("Process Text", visible=False) 
-
-    # --- Connect inputs and outputs within gr.Blocks ---
-    interface = gr.Interface(
-        fn=process_video,  
-        inputs=[video, sitting, hands, location, screen],
-        outputs=[process_button, video_description], 
-        examples=examples,
-        title=title,
-        description=description,
-        article=article,
-        allow_flagging="never",
-    )
-    interface.launch(debug=False)
-
-    # --- Button click event handler ---
-    process_button.click(fn=process_and_display_text, outputs=json_output)# Click event for the "Process Text" button
-
-
 interface.launch(debug=False)