Update app.py

app.py

@@ -1,29 +1,29 @@
-import gradio as gr
-from transformers import AutoProcessor, AutoModel
-
-# Step 1: Load the model and processor
-model_name = "facebook/VFusion3D"
-processor = AutoProcessor.from_pretrained(model_name)
-model = AutoModel.from_pretrained(model_name)
-
-# Step 2: Define the function to process inputs and get predictions
-def predict(input_text):
-    # Convert input into a format the model understands
-    inputs = processor(inputs=input_text, return_tensors="pt")
-    # Get model predictions
-    outputs = model(**inputs)
-    # Return results (adjust based on the model's output format)
-    return outputs.logits.tolist()
-
-# Step 3: Build the Gradio interface
-interface = gr.Interface(
-    fn=predict,
-    inputs="text",  # Change this based on model requirements
-    outputs="text",  # Adjust output format as needed
-    title="VFusion3D Deployment",
-    description="A demo application."
-)
-
-# Step 4: Launch the app
-if __name__ == "__main__":
-    interface.launch()
+import gradio as gr
+from transformers import AutoProcessor, AutoModel
+
+# Model name
+model_name = "facebook/VFusion3D"
+
+# Load processor and model with trusted code
+processor = AutoProcessor.from_pretrained(model_name, trust_remote_code=True)
+model = AutoModel.from_pretrained(model_name, trust_remote_code=True)
+
+# Define prediction function
+def predict(input_text):
+    # Convert input into a format the model understands
+    inputs = processor(inputs=input_text, return_tensors="pt")
+    outputs = model(**inputs)
+    return outputs.logits.tolist()
+
+# Gradio interface
+interface = gr.Interface(
+    fn=predict,
+    inputs="text",
+    outputs="text",
+    title="VFusion3D Deployment",
+    description="A demo for facebook/VFusion3D model."
+)
+
+# Launch the app
+if __name__ == "__main__":
+    interface.launch()