ECharts散点图+线性回归(机器学习)

有一份酒店数据，

 收购(购入)酒店的价格 和 酒店面积 的关系 

 ECharts 效果展示 :

 不加线性回归代码:

from flask import Flask,render_template
import pandas as pd
import settings

app = Flask(__name__)
app.config.from_object(settings)

file = pd.read_csv(r"D:\hotel\hotels.csv")     # 文件存放地址

# 酒店收入价格与酒店面积的关系
# @app.route('/')
def hotel():
    hh = file.loc[:,["Price in Millions","Square Meter"]].rename(columns={"Price in Millions":"price","Square Meter":"area"})
    data = []
    for idx,row in hh.iterrows():
        data.append([row.area,row.price])
    print(data)

if __name__ == '__main__':
    hotel()
    # app.run()

打印结果(传入ECharts的结果)

 加入线性回归:

# 从sklearn库里导入
from sklearn.linear_model import LinearRegression

# 创建对象
lr = LinearRegression()

# 对上面的data进行后续操作
# data -> 准备好传入Echarts画散点图的格式
# 获取 x , y
data_x = [[xx[0]] for xx in data]
data_y = [[xx[1]] for xx in data]

lr.fit(data_x,data_y)

# 对 data_x 进行预测
temp = lr.predict(data_x)

# 创建一个空列表用于存放线性回归的数据
data_predict = []

# 遍历往创建的空列表存放数据
for i in range(len(temp)):
    data_predict.append([data_x[i][0],temp[i][0]])

合并两个部分 :

from flask import Flask,render_template
from sklearn.linear_model import LinearRegression
import pandas as pd
import settings

app = Flask(__name__)
app.config.from_object(settings)

file = pd.read_csv(r"D:\hotel\hotels.csv")
lr = LinearRegression()

# 酒店收入价格与酒店面积的关系
@app.route('/')
def hotel():
    hh = file.loc[:,["Price in Millions","Square Meter"]].rename(columns={"Price in Millions":"price","Square Meter":"area"})
    data = []
    for idx,row in hh.iterrows():
        data.append([row.area,row.price])
    # print(data)

    data_x = [[xx[0]] for xx in data]
    data_y = [[xx[1]] for xx in data]
    lr.fit(data_x,data_y)
    temp = lr.predict(data_x)
    # print(temp)
    data_y_predict = []
    for i in range(len(temp)):
        data_y_predict.append([data_x[i][0],temp[i][0]])
    print(data_y_predict)

    return render_template('hotel.html',data=data,data_y_predict=data_y_predict)


if __name__ == '__main__':
    # hotel()
    app.run()

最后在 画图的时候加上 一条线性回归的线就可了