求助： Python flask 应用内存一直在增加

2 个版本，都没有解决内存持续增加，由于内存一直增加，最后因为小鸡内存不足而被杀掉。

这是一个加载训练后的模型，通过网络传入预测参数，然后返回预测 json 结果。

可能同时发起（并发）几十个请求。

求大佬帮忙看看问题出在哪里，谢谢。

这是 chatgpt 4.0 给的版本

from flask import Flask, request, jsonify import pickle import os import psutil import pandas as pd app = Flask(__name__) class SingletonModel: _instance = None def __new__(cls): if cls._instance is None: print("Creating Singleton Instance") cls._instance = super(SingletonModel, cls).__new__(cls) modelName = "xgboost_model-k.pkl" with open(modelName, "rb") as pkl_file: loaded_data = pickle.load(pkl_file) cls._instance.model = loaded_data['model'] cls._instance.scaler = loaded_data['scaler'] cls._instance.label_encoder = loaded_data['label_encoder'] cls._instance.feature_names = ['shortAvg','longAvg','volatility','diff'] return cls._instance resources = SingletonModel() model = resources.model scaler = resources.scaler label_encoder = resources.label_encoder @app.route('/predict', methods=['POST']) def predict(): global model, scaler, label_encoder data = request.json['input'] df = pd.DataFrame([data], columns=resources.feature_names) scaled_data = scaler.transform(df) prediction = model.predict(scaled_data) label_prediction = label_encoder.inverse_transform(prediction) return jsonify([label_prediction[0]]) if __name__ == '__main__': app.run(port=6601,debug=True) 

这是 Claude 给的版本

 import asyncio from flask import Flask, request, jsonify import pickle import pandas as pd app = Flask(__name__) # 模型相关全局变量 model = None scaler = None label_encoder = None async def load_model(): global model, scaler, label_encoder,feature_names if not model: with open('xgboost_model-k.pkl', 'rb') as f: loaded_data = pickle.load(f) model = loaded_data['model'] scaler = loaded_data['scaler'] label_encoder = loaded_data['label_encoder'] feature_names = ['shortAvg','longAvg','volatility','diff'] async def predict(data): await load_model() df = pd.DataFrame([data], columns=feature_names) scaled_data = scaler.transform(df) prediction = model.predict(scaled_data) label_prediction = label_encoder.inverse_transform(prediction) return label_prediction[0] @app.route('/predict', methods=['POST']) async def predict_handler(): data = request.json['input'] result = await asyncio.gather(predict(data)) return jsonify(result) if __name__ == '__main__': app.run(port=6601,debug=False)