补充readme

flask_batch.py

@@ -1,3 +1,5 @@
+import os
+os.environ["CUDA_VISIBLE_DEVICES"] = "0"
 from flask import Flask, jsonify
 from flask import request
 from transformers import pipeline

predict.py

@@ -1,5 +1,4 @@
 import time
-
 from vllm import LLM, SamplingParams
 
 prompts = [