【demo】python 按照标签分组取前N行

强烈推介IDEA2021.1.3破解激活，IntelliJ IDEA 注册码，2021.1.3IDEA 激活码

1. import pandas as pd

2. train_data = pd.read_csv(filepath_or_buffer='E:\\tmp\\2\\tagging.tsv', sep='\t',encoding='utf-8',index_col=0,header=None,

3. names=['label', 'textA', 'textB'])

4. train_data.drop_duplicates(inplace=True)

5. df = pd.DataFrame(train_data)

6. grouped = df.groupby(['label']).head(3000)

7. grouped = grouped.sort_values(['label'], ascending=True)

8. #

9. print(grouped)

10. grouped.to_csv('E:\\tmp\\2\\tagging2.tsv',

11. sep='\t',

12. encoding='utf-8',

13. header=None)