Spaces:
Runtime error
Runtime error
File size: 2,352 Bytes
b10cf74 |
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 |
import gradio as gr
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from huggingface_hub import snapshot_download
# 🔹 Download & load the model from Hugging Face
model_name = "HyperX-Sen/Qwen-2.5-7B-Reasoning"
model_path = snapshot_download(repo_id=model_name, repo_type="model")
# 🔹 Load the model & tokenizer
model = AutoModelForCausalLM.from_pretrained(model_path, torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_path)
# 🔹 System prompt
SYSTEM_PROMPT = """
Respond in the following format:
<reasoning>
...
</reasoning>
<answer>
...
</answer>
"""
# 🔹 Function to generate response
def chat_response(user_input, top_p, top_k, temperature, max_length):
messages = [
{"role": "system", "content": f"{SYSTEM_PROMPT}"},
{"role": "user", "content": user_input}
]
# 🔹 Format & tokenize input
input_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(input_text, return_tensors="pt").to(model.device)
# 🔹 Generate response
with torch.no_grad():
output = model.generate(
**inputs,
max_length=max_length,
do_sample=True,
top_p=top_p,
top_k=top_k,
temperature=temperature
)
# 🔹 Decode output
response = tokenizer.decode(output[0], skip_special_tokens=True)
return response
# 🔹 Gradio UI
with gr.Blocks() as demo:
gr.Markdown("# 🤖 Qwen-2.5-7B-Reasoning Chatbot")
with gr.Row():
chatbot = gr.Textbox(label="Model Response", lines=8, interactive=False)
with gr.Row():
user_input = gr.Textbox(label="Your Prompt", placeholder="Ask me anything...", lines=2)
with gr.Accordion("🔧 Advanced Settings", open=False):
top_p = gr.Slider(0.1, 1.0, value=0.9, label="Top-p")
top_k = gr.Slider(1, 100, value=50, label="Top-k")
temperature = gr.Slider(0.1, 1.5, value=0.7, label="Temperature")
max_length = gr.Slider(128, 1024, value=512, label="Max Length")
with gr.Row():
submit_button = gr.Button("Generate Response")
submit_button.click(chat_response, inputs=[user_input, top_p, top_k, temperature, max_length], outputs=[chatbot])
# 🔹 Launch the Gradio app
demo.launch()
|