init

app.py

@@ -117,7 +117,7 @@ def camera_understanding(image_src, question, seed, progress=gr.Progress(track_t
             lat_img = fig_to_image(fig)
         plt.close(fig)
 
-    return text, up_img, lat_img
+    return text#, up_img, lat_img
 
 
 @torch.inference_mode()
@@ -214,8 +214,8 @@ with gr.Blocks(css=css) as demo:
         understanding_button = gr.Button("Chat")
         understanding_output = gr.Textbox(label="Response")
         
-        camera1 = gr.Gallery(label="Camera Maps", columns=1, rows=1)
-        camera2 = gr.Gallery(label="Camera Maps", columns=1, rows=1)
+        #camera1 = gr.Gallery(label="Camera Maps", columns=1, rows=1)
+        #camera2 = gr.Gallery(label="Camera Maps", columns=1, rows=1)
 
         with gr.Accordion("Advanced options", open=False):
             und_seed_input = gr.Number(label="Seed", precision=0, value=42)
@@ -242,7 +242,7 @@ with gr.Blocks(css=css) as demo:
     understanding_button.click(
         camera_understanding,
         inputs=[image_input, und_seed_input],
-        outputs=[understanding_output, camera1, camera2]
+        outputs=[understanding_output]#, camera1, camera2]
     )
 
 demo.launch(share=True)