1

2

3

df_dropped = df.dropna()

print("\n删除缺失值后:")

print(df_dropped)

1

2

3

4

5

6

7

df_filled = df.fillna({

    'title': 'Unknown',

    'author': 'Unknown Author',

    'price': df['price'].mean()

})

print("\n填充缺失值后:")

print(df_filled)

1

2

3

df_ffill = df.fillna(method='ffill')

print("\n前向填充缺失值后:")

print(df_ffill)

1

2

3

df_bfill = df.fillna(method='bfill')

print("\n后向填充缺失值后:")

print(df_bfill)

1

2

3

df_dropped_cols = df.dropna(axis=1, thresh=len(df) * 0.5) 

print("\n删除缺失率高的列后:")

print(df_dropped_cols)

1

2

3

df_deduplicated = df.drop_duplicates()

print("\n删除重复值后:")

print(df_deduplicated)

1

2

3

df['is_duplicate'] = df.duplicated()

print("\n标记重复值后:")

print(df)

1

2

3

df['price'] = df['price'].apply(lambda x: x if 0 <= x <= 100 else df['price'].mean())

print("\n替换异常值后:")

print(df)

1

2

3

df['is_outlier'] = df['price'].apply(lambda x: 1 if x < 0 or x > 100 else 0)

print("\n标记异常值后:")

print(df)

1

2

3

df['price'] = df['price'].astype(int)

print("\n转换为整数后:")

print(df)

1

2

3

df['date'] = pd.to_datetime(df['date'], errors='coerce')

print("\n转换为日期类型后:")

print(df)

1

2

3

df['title'] = df['title'].str.strip()

print("\n去掉两端空格后:")

print(df)

1

2

3

df['title'] = df['title'].str.replace('[^a-zA-Z0-9\s]', '', regex=True)

print("\n替换特定字符后:")

print(df)

1

2

3

df['title'] = df['title'].str.lower()

print("\n转换为小写后:")

print(df)

1

2

3

grouped = df.groupby('author')['price'].mean()

print("\n按作者分组的平均价格:")

print(grouped)

1

2

3

4

5

bins = [0, 10, 20, 30]

labels = ['低', '中', '高']

df['price_level'] = pd.cut(df['price'], bins=bins, labels=labels, right=False)

print("\n按价格分箱后:")

print(df)

1

2

3

4

5

from sklearn.preprocessing import MinMaxScaler

scaler = MinMaxScaler()

df['price_scaled'] = scaler.fit_transform(df[['price']])

print("\n归一化后的数据:")

print(df)