File size: 2,213 Bytes
cfba9db a7aa06f cfba9db c654a7e cfba9db a7aa06f 2cfcd90 a7aa06f cfba9db 2cfcd90 26c5e44 2cfcd90 26c5e44 cfba9db c654a7e cfba9db c654a7e cfba9db c654a7e cfba9db 26c5e44 cfba9db 26c5e44 |
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 |
import logging
from flask import Flask, request, jsonify
from transformers import pipeline
import gradio as gr
import os
import torch
from huggingface_hub import login
# Cấu hình logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
# Đăng nhập Hugging Face
try:
login(token=os.getenv("HF_TOKEN"))
logging.info("Logged in to Hugging Face Hub successfully")
except Exception as e:
logging.error(f"Failed to login to Hugging Face Hub: {e}")
raise
# Load mô hình
logging.info("Loading nguyenvulebinh/vi-mrc-base...")
try:
qa_pipeline = pipeline(
"question-answering",
model="nguyenvulebinh/vi-mrc-base",
device=0 if torch.cuda.is_available() else -1
)
logging.info("Model loaded successfully")
except Exception as e:
logging.error(f"Failed to load model: {e}")
raise
# Hàm xử lý cho Gradio và API
def gradio_answer(question, context):
result = qa_pipeline(question=question, context=context)
return result["answer"]
# Tạo Flask app để thêm endpoint API
app = Flask(__name__)
@app.route("/api/answer", methods=["POST"])
def answer():
try:
data = request.json
question = data.get("question")
context = data.get("context")
logging.info(f"Received request - Question: {question}, Context: {context[:200]}...")
if not question or not context:
logging.error("Missing question or context")
return jsonify({"error": "Missing question or context"}), 400
result = qa_pipeline(question=question, context=context)
logging.info(f"Response - Answer: {result['answer']}")
return jsonify({"answer": result["answer"]})
except Exception as e:
logging.error(f"API error: {e}")
return jsonify({"error": str(e)}), 500
# Tạo Gradio interface
iface = gr.Interface(
fn=gradio_answer,
inputs=["text", "text"],
outputs="text",
title="AgriBot: Hỏi đáp nông nghiệp",
description="Nhập câu hỏi và ngữ cảnh để nhận câu trả lời về nông nghiệp."
)
# Thêm Flask app vào Gradio
iface.launch(server_name="0.0.0.0", server_port=7860, app=app) |