1.加载数据与数据处理
如何建立csv文件
①(.csv文件)
import pandas as pd
data = pd.read_csv(r'xxx.csv')
②(.mat文件)
from scipy.io import loadmat#用于加载mat文件
data_m= loadmat("xxx.mat")
print(data_m.keys())#其中有很多key,注意找自己数据所在的那个
data_load = data_m["ML_load"]#我需要的是这个,这是一个2998*10的矩阵
data_load = pd.DataFrame(data_load,columns=("load1","load2","load3","load4","load5","load6","load7","load8","load9","load10",))#给每列加上一个名字或者叫特征名,此时data_load就是一个dataframe类型了。
print(data_load.head())
print(data_load.shape)
2.数据集与训练集的划分
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0,test_size=0.2)#随机划分训练集,测试集
print(X_train.shape,y_train.shape,X_test.shape,y_test.shape)