One-hot encoding 数据处理

最新推荐文章于 2025-05-17 12:46:16 发布

原创最新推荐文章于 2025-05-17 12:46:16 发布 · 484 阅读

CC 4.0 BY-SA版权

本文介绍了一个使用Python进行OneHot编码的实际案例。通过处理CSV文件中的多类别特征数据，并利用sklearn库中的LabelBinarizer实现OneHot编码转换，将文本类别特征转化为数值型特征，便于后续的数据分析和机器学习应用。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

import csv
import os
import shutil
import codecs
import pandas as pd
import numpy as np

from sklearn.preprocessing import OneHotEncoder
from sklearn.preprocessing import LabelEncoder
from sklearn.preprocessing import LabelBinarizer
from sklearn.preprocessing import MultiLabelBinarizer

dir_name = ‘C:\Users\Thuang6\Desktop\MaxWellData\OneHot\csv_to_csv.csv’

path = os.chdir(‘C:\Users\Thuang6\Desktop\MaxWellData\OneHot’)
df = pd.read_csv(dir_name,names=[‘Time’,’Process’,’Component’,’Operation’,’Action’,’Control’,’Category’,’Context’],index_col = False)
df = df.fillna(value= ‘NULL’)
process = LabelBinarizer().fit_transform(df[‘Process’])
print(process)

component = LabelBinarizer().fit_transform(df[‘Component’])
print(component)

operation = LabelBinarizer().fit_transform(df[‘Operation’])
print(operation)

action = LabelBinarizer().fit_transform(df[‘Action’])
print(action)

control = LabelBinarizer().fit_transform(df[‘Control’])
print(control)

category = LabelBinarizer().fit_transform(df[‘Category’])
print(category)

final_output = np.hstack((process,component,operation,action,control,category))

print(final_output)

final_split = np.vsplit(final_output,21)

print(final_split)

print(np.shape(final_split))

print(“nihao”)
d = []
for i in range(21):
a = final_split[i]
#print(a)
b = np.ndarray.flatten(a)
c = b.tolist()
d.append(c)
#print(d)
#print(len(d))
#print(type(b))