ML之FE：在模型訓(xùn)練中，僅需兩行代碼實(shí)現(xiàn)切分訓(xùn)練集和測(cè)試集并分離特征與標(biāo)簽

?

?

?

?

目錄

僅需兩行代碼實(shí)現(xiàn)切分訓(xùn)練集和測(cè)試集并分離特征與標(biāo)簽

輸出結(jié)果

實(shí)習(xí)代碼

---

?

?

?

僅需兩行代碼實(shí)現(xiàn)切分訓(xùn)練集和測(cè)試集并分離特征與標(biāo)簽

輸出結(jié)果

name object ID object age object age02 int64 age03 object born datetime64[ns] sex object hobbey object money float64 weight float64 test01 float64 test02 float64 dtype: objectname ID age age02 age03 born sex hobbey money weight \ 0 Bob 1 NaN 14 14 NaT 男 打籃球 200.0 140.5 1 LiSa 2 28 26 26 1990-01-01 女 打羽毛球 240.0 120.8 2 Mary 38 24 24 1980-01-01 女 打乒乓球 290.0 169.4 3 Alan None 6 6 NaT None 300.0 155.6 test01 test02 0 1.000000 1.000000 1 2.123457 2.123457 2 3.123457 3.123457 3 4.123457 4.123457 0 140.5 1 120.8 2 169.4 Name: weight, dtype: float64name ID age age02 age03 born sex hobbey money weight test01 \ 3 Alan None 6 6 NaT None 300.0 155.6 4.123457 test02 3 4.123457

?

?

實(shí)習(xí)代碼

import pandas as pd import numpy as npcontents={"name": ['Bob', 'LiSa', 'Mary', 'Alan'],"ID": [1, 2, ' ', None], # 輸出 NaN"age": [np.nan, 28, 38 , '' ], # 輸出 "age02": [14, 26, 24 , 6], "age03": [14, '26', '24' , '6'], "born": [pd.NaT, pd.Timestamp("1990-01-01"), pd.Timestamp("1980-01-01"), ''], # 輸出 NaT"sex": ['男', '女', '女', None,], # 輸出 None"hobbey":['打籃球', '打羽毛球', '打乒乓球', '',], # 輸出 "money":[200.0, 240.0, 290.0, 300.0], # 輸出"weight":[140.5, 120.8, 169.4, 155.6], # 輸出"test01":[1, 2.123456789, 3.123456781011126, 4.123456789109999], # 輸出"test02":[1, 2.123456789, 3.123456781011126, 4.123456789109999], # 輸出} data_frame = pd.DataFrame(contents) # data_frame.to_excel("data_Frame.xls") print(data_frame.dtypes) print(data_frame)# ML之FE：在模型訓(xùn)練中，僅需兩行代碼實(shí)現(xiàn)切分訓(xùn)練集和測(cè)試集并分離特征與標(biāo)簽 train_test_split_Index=3 label_col='weight' train_X = data_frame[:train_test_split_Index] train_y = data_frame[:train_test_split_Index][label_col] test_X = data_frame[train_test_split_Index:] test_y = data_frame[train_test_split_Index:][label_col] print(train_y) print(test_X)

?

?

?

?

?

?

?

?

?

總結(jié)

以上是生活随笔為你收集整理的ML之FE：在模型训练中，仅需两行代码实现切分训练集和测试集并分离特征与标签的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問(wèn)題。

如果覺(jué)得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。