如何在Python中进行数据预处理和特征工程

import pandas as pd

# 读取CSV文件
data = pd.read_csv('data.csv')

# 检查缺失值
data.isnull().sum()

# 填充缺失值
data['column_name'].fillna(data['column_name'].mean(), inplace=True)

# 检查异常值
data['column_name'].describe()

# 替换异常值
data['column_name'].replace({-999: np.nan}, inplace=True)

# 删除重复值
data.drop_duplicates(inplace=True)

from sklearn.feature_selection import VarianceThreshold

# 设置方差阈值
selector = VarianceThreshold(threshold=0.1)

# 进行特征选择
selected_features = selector.fit_transform(data)

# 计算特征之间的相关系数
correlation_matrix = data.corr()

# 筛选相关性较高的特征
highly_correlated_features = correlation_matrix[correlation_matrix > 0.8].dropna(axis=0).index
selected_features = data[highly_correlated_features]

from sklearn.feature_extraction.text import CountVectorizer

# 实例化文本特征提取器
text_vectorizer = CountVectorizer()

# 提取文本特征
text_features = text_vectorizer.fit_transform(data['text_column'])

import cv2

# 读取图像
image = cv2.imread('image.jpg')

# 提取图像特征
image_features = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

# 转换时间格式
data['timestamp'] = pd.to_datetime(data['timestamp'])

# 提取时间序列特征
data['year'] = data['timestamp'].dt.year
data['month'] = data['timestamp'].dt.month