File size: 2,242 Bytes
7f692e6
3382226
 
7f692e6
3382226
 
 
 
7f692e6
3382226
 
 
 
 
 
 
 
7f692e6
3382226
 
 
7f692e6
 
3382226
7f692e6
3382226
7f692e6
 
3382226
 
 
 
 
 
 
 
7f692e6
3382226
 
7f692e6
3382226
 
 
 
 
 
 
7f692e6
 
 
3382226
7f692e6
 
 
 
3382226
7f692e6
 
 
 
 
 
3382226
7f692e6
 
 
 
 
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
import gradio as gr
from unsloth import FastLanguageModel
from transformers import TextStreamer

# Load the model and tokenizer locally
max_seq_length = 2048
dtype = None
model_name_or_path = "michailroussos/model_llama_8d"

# Load model and tokenizer using unsloth
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name=model_name_or_path,
    max_seq_length=max_seq_length,
    dtype=dtype,
    load_in_4bit=True,
)
FastLanguageModel.for_inference(model)  # Enable optimized inference

# Define the response function
def respond(message, history, system_message, max_tokens, temperature, top_p):
    # Build the chat message history
    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})
    
    # Tokenize the input messages
    inputs = tokenizer.apply_chat_template(
        messages,
        tokenize=True,
        add_generation_prompt=True,  # Required for generation
        return_tensors="pt",
    ).to("cuda")

    # Initialize a TextStreamer for streaming output
    text_streamer = TextStreamer(tokenizer, skip_prompt=True)

    # Generate the model's response
    response = ""
    for output in model.generate(
        input_ids=inputs,
        streamer=text_streamer,
        max_new_tokens=max_tokens,
        use_cache=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = tokenizer.decode(output, skip_special_tokens=True)
        response += token
        yield response


# Define the Gradio 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__":
    demo.launch()