File size: 1,153 Bytes
ef9cde4
 
 
 
 
 
 
4da81f8
ef9cde4
 
 
 
 
 
4da81f8
ef9cde4
 
 
4da81f8
 
ef9cde4
 
 
 
 
 
 
 
4da81f8
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
import os
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# 讟讜注谉 讗转 讛诪讜讚诇 讜讛-tokenizer
tokenizer = AutoTokenizer.from_pretrained('dicta-il/dictalm-7b-instruct')
model = AutoModelForCausalLM.from_pretrained('dicta-il/dictalm-7b-instruct', trust_remote_code=True)

# 讛讙讚专转 讛驻讜谞拽爪讬讛 诇爪'讗讟 注诐 讛诪讜讚诇
def chat_with_model(prompt):
    model.eval()
    with torch.inference_mode():
        kwargs = dict(
            inputs=tokenizer(prompt, return_tensors='pt').input_ids,
            do_sample=True,
            top_k=50,
            top_p=0.95,
            temperature=0.5,  # 讛讜专讚转 讛讟诪驻专讟讜专讛 诇讛拽讟谞转 讛讗拽专讗讬讜转
            max_length=50,  # 讛拽讟谞转 讛诪拽住讬诪讜诐 诇诪住驻专 拽讟谉 讬讜转专
            min_new_tokens=5
        )
        output = model.generate(**kwargs)
        response_text = tokenizer.batch_decode(output, skip_special_tokens=True)[0]
    return response_text

# 讬爪讬专转 诪诪砖拽 注诐 Gradio
interface = gr.Interface(fn=chat_with_model, inputs="text", outputs="text", title="Chat with DictaLM Model")
interface.launch()