Spaces:
Sleeping
Sleeping
File size: 3,375 Bytes
2ac4330 6a1d4c8 2ac4330 87569bb 6a1d4c8 2ac4330 6a1d4c8 2ac4330 6a1d4c8 2ac4330 6a1d4c8 87569bb 2ac4330 6a1d4c8 2ac4330 1aabada 40a939a 2ac4330 87569bb 2ac4330 6a1d4c8 |
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 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 |
import gradio as gr
from huggingface_hub import InferenceClient
"""
For more information on `huggingface_hub` Inference API support, please check the docs:
https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
def respond(message, history, token, model, system_message, max_tokens, temperature, top_p):
"""
Handle chat responses using the Hugging Face Inference API.
Parameters:
- message: The user's current message.
- history: List of previous user-assistant message pairs.
- token: HF API token for authentication.
- model: Model name to use for inference.
- system_message: System prompt to initialize the chat.
- max_tokens: Maximum number of tokens to generate.
- temperature: Sampling temperature.
- top_p: Top-p (nucleus) sampling parameter.
Yields:
- Incremental responses for streaming in the chat interface.
"""
# Check for missing token
if not token:
yield "Please provide an HF API Token."
return
# Use default model if none provided
if not model:
model = "meta-llama/Llama-3.1-8B-Instruct"
# Initialize the InferenceClient
try:
client = InferenceClient(model=model, token=token)
except Exception as e:
yield f"Error initializing client: {str(e)}"
return
# Build the message history, starting with the system message
messages = [{"role": "system", "content": system_message}]
for val in history:
if val[0]: # User message
messages.append({"role": "user", "content": val[0]})
if val[1]: # Assistant message
messages.append({"role": "assistant", "content": val[1]})
messages.append({"role": "user", "content": message})
response = ""
# Make the API call with streaming
try:
for message in client.chat_completion(
messages,
max_tokens=max_tokens,
stream=True,
temperature=temperature,
top_p=top_p,
):
# Check for non-empty content in the delta
if message.choices and message.choices[0].delta.content is not None:
token = message.choices[0].delta.content
response += token
yield response
except Exception as e:
yield f"Error during API call: {str(e)}"
# Define input components
token_input = gr.Textbox(type="password", label="HF API Token")
model_input = gr.Textbox(label="Model Name", value="HuggingFaceH4/zephyr-7b-beta")
"""
For information on how to customize the ChatInterface, peruse the gradio docs:
https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
title="Chat Interface to test HF Models with a HF TOKEN",
description="Enter your HF Token To Acces the Models",
respond,
additional_inputs=[
token_input,
model_input,
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__":
demo.launch() |