Spaces:
Sleeping
Sleeping
File size: 1,822 Bytes
12d1a48 f364d75 12d1a48 f364d75 12d1a48 f364d75 12d1a48 f364d75 12d1a48 f364d75 12d1a48 f364d75 12d1a48 f364d75 |
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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 |
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
# Load DialoGPT model and tokenizer
model_name = "microsoft/DialoGPT-large"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
# Respond function for Gradio interface
def respond(
message,
history: list[tuple[str, str]],
system_message,
max_tokens,
temperature,
top_p,
):
# Format the chat history for the DialoGPT model
full_conversation = ""
for user_msg, bot_msg in history:
if user_msg:
full_conversation += f"User: {user_msg}\n"
if bot_msg:
full_conversation += f"DialoGPT: {bot_msg}\n"
full_conversation += f"User: {message}\nDialoGPT:"
# Tokenize input and generate response
inputs = tokenizer.encode(full_conversation, return_tensors="pt")
outputs = model.generate(
inputs,
max_length=max_tokens,
temperature=temperature,
top_p=top_p,
pad_token_id=tokenizer.eos_token_id,
)
response = tokenizer.decode(outputs[:, inputs.shape[-1] :][0], skip_special_tokens=True)
return response
# Gradio Chat Interface
demo = gr.ChatInterface(
respond,
additional_inputs=[
gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
gr.Slider(
minimum=0.1,
maximum=1.0,
value=0.95,
step=0.05,
label="Top-p (nucleus sampling)",
),
],
)
if __name__ == "__main__":
# Launch the Gradio app with API enabled
demo.launch(enable_api=True)
|