algorithm_platform/models_processing/model_predict/model_prediction_lstm.py

from flask import Flask,request
import time
import logging
import traceback
import numpy as np
from itertools import chain
from common.database_dml import get_data_from_mongo,insert_data_into_mongo,get_h5_model_from_mongo,get_scaler_model_from_mongo
app = Flask('model_prediction_lstm——service')


# 创建时间序列数据
def create_sequences(data_features,data_target,time_steps):
    X, y = [], []
    if len(data_features)<time_steps:
        print("数据长度不能比时间步长小！")
        return np.array(X), np.array(y)
    else:
        for i in range(len(data_features) - time_steps+1):
            X.append(data_features[i:(i + time_steps)])
            if len(data_target)>0:
                y.append(data_target[i + time_steps -1])
        return np.array(X), np.array(y)


def model_prediction(df,args):
    features, time_steps, col_time, model_name =  str_to_list(args['features']), int(args['time_steps']),args['col_time'],args['model_name']
    feature_scaler,target_scaler = get_scaler_model_from_mongo(args)
    df = df.fillna(method='ffill').fillna(method='bfill').sort_values(by=col_time)
    scaled_features = feature_scaler.transform(df[features])
    X_predict, _ = create_sequences(scaled_features, [], time_steps)
    # 加载模型时传入自定义损失函数
    # model = load_model(f'{farmId}_model.h5', custom_objects={'rmse': rmse})
    model = get_h5_model_from_mongo(args)
    y_predict = list(chain.from_iterable(target_scaler.inverse_transform([model.predict(X_predict).flatten()])))
    result = df[-len(y_predict):]
    result['predict'] = y_predict
    result['model'] = model_name
    return result

def str_to_list(arg):
    if arg == '':
        return []
    else:
        return arg.split(',')

@app.route('/model_prediction_lstm', methods=['POST'])
def model_prediction_lstm():
    # 获取程序开始时间  
    start_time = time.time()  
    result = {}
    success = 0
    print("Program starts execution!")
    try:
        args = request.values.to_dict()
        print('args',args)
        logger.info(args)
        power_df = get_data_from_mongo(args)
        model = model_prediction(power_df,args)
        insert_data_into_mongo(model,args)
        success = 1
    except Exception as e:
        my_exception = traceback.format_exc()
        my_exception.replace("\n","\t")
        result['msg'] = my_exception
    end_time = time.time() 
   
    result['success'] = success
    result['args'] = args
    result['start_time'] = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(start_time))
    result['end_time'] = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(end_time))
    print("Program execution ends!")
    return result


if __name__=="__main__":  
    print("Program starts execution!")
    logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
    logger = logging.getLogger("model_prediction_lstm log")
    from waitress import serve
    serve(app, host="0.0.0.0", port=10097)
    print("server start!")