File size: 1,115 Bytes
b21dbf7 5be10eb b21dbf7 5be10eb b21dbf7 5be10eb b21dbf7 | 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 | from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
import gradio as gr
BASE_MODEL = "meta-llama/Llama-2-7b-hf"
ADAPTER = "thangduong0509/vinallama-peft-7b-math-solver"
# Load base model
tokenizer = AutoTokenizer.from_pretrained(BASE_MODEL, trust_remote_code=True)
base_model = AutoModelForCausalLM.from_pretrained(
BASE_MODEL,
device_map="auto",
torch_dtype=torch.float16,
trust_remote_code=True
)
# Load PEFT adapter
model = PeftModel.from_pretrained(base_model, ADAPTER)
model.eval()
def solve_math(prompt):
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
with torch.no_grad():
outputs = model.generate(**inputs, max_new_tokens=128, do_sample=True, temperature=0.7)
return tokenizer.decode(outputs[0], skip_special_tokens=True)
gr.Interface(
fn=solve_math,
inputs=gr.Textbox(lines=3, placeholder="Nhập đề Toán tiểu học..."),
outputs="text",
title="🧮 ViLLama Math Solver",
description="Giải toán tiểu học bằng mô hình LLaMA-2 7B fine-tune với PEFT"
).launch()
|