File size: 811 Bytes
0899c0c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
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()