机器学习模型部署远程服务功能

框架  flask

example

1.sklearn 训练模型并保持

import numpy as np

from sklearn.linear_model import LinearRegression

from sklearn.utils import check_random_state

from sklearn.externals import joblib

n = 100

x = np.arange(n)

rs = check_random_state(0)

y = rs.randint(-50, 50, size=(n,)) + 50. * np.log1p(np.arange(n))

 

lr = LinearRegression()

lr.fit(x[:, np.newaxis], y)

joblib.dump(lr,'lr_model.pkl')

 

2.训练好的模型提供远程服务

import numpy as np

from flask import Flask

from flask import request

from flask import jsonify

from sklearn.externals import joblib

#导入模型

model = joblib.load('lr_model.pkl')

app = Flask(__name__)

@app.route('/',methods=['POST','GET'])

def output_data():

    text=request.args.get('inputdata')

    if text:

        temp =  [float(x) for x in text.split(',')]

        temp = np.array(temp).reshape((1, -1))

        ouputdata = model.predict(temp)

        return jsonify(str(ouputdata[0]))

 

if __name__ == '__main__':

    app.config['JSON_AS_ASCII'] = False

    app.run(host='0.0.0.0',port=5003)  # 127.0.0.1 #指的是本地ip，0,0,0,0指支持远程访问

    print('运行结束')

 

3.测试

test.py

import requests

base = 'http://127.0.0.1:5003/?inputdata=5'   #ip地址换成启动服务的机器地址

response = requests.get(base)

answer = response.json()

print('预测结果',answer)