Spaces:
Sleeping
Sleeping
predict_text
Browse files
app.py
CHANGED
@@ -55,11 +55,29 @@ def predict(image, text):
|
|
55 |
response = processor.decode(outputs[0], skip_special_tokens=True)
|
56 |
return response
|
57 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
58 |
# Define the Gradio interface
|
59 |
interface = gr.Interface(
|
60 |
-
fn=
|
61 |
inputs=[
|
62 |
-
gr.Image(type="pil", label="Image Input"), # Image input with label
|
63 |
gr.Textbox(label="Text Input") # Textbox input with label
|
64 |
],
|
65 |
outputs=gr.Textbox(label="Generated Response"), # Output with a more descriptive label
|
|
|
55 |
response = processor.decode(outputs[0], skip_special_tokens=True)
|
56 |
return response
|
57 |
|
58 |
+
def predict_text(text):
|
59 |
+
# Prepare the input messages
|
60 |
+
messages = [{"role": "user", "content": [{"type": "text", "text": txt}]}]
|
61 |
+
|
62 |
+
# Create the input text using the processor's chat template
|
63 |
+
input_text = processor.apply_chat_template(messages, add_generation_prompt=True)
|
64 |
+
|
65 |
+
# Process the inputs and move to the appropriate device
|
66 |
+
# inputs = processor(image, input_text, return_tensors="pt").to(device)
|
67 |
+
inputs = processor(text=text, return_tensors="pt").to("cuda")
|
68 |
+
# Generate a response from the model
|
69 |
+
outputs = model.generate(**inputs, max_new_tokens=250)
|
70 |
+
|
71 |
+
# Decode the output to return the final response
|
72 |
+
response = processor.decode(outputs[0], skip_special_tokens=True)
|
73 |
+
return response
|
74 |
+
|
75 |
+
|
76 |
# Define the Gradio interface
|
77 |
interface = gr.Interface(
|
78 |
+
fn=predict_text,
|
79 |
inputs=[
|
80 |
+
# gr.Image(type="pil", label="Image Input"), # Image input with label
|
81 |
gr.Textbox(label="Text Input") # Textbox input with label
|
82 |
],
|
83 |
outputs=gr.Textbox(label="Generated Response"), # Output with a more descriptive label
|