【demo】python 按照标签分组取前N行

import pandas as pd
train_data = pd.read_csv(filepath_or_buffer='E:\\tmp\\2\\tagging.tsv', sep='\t',encoding='utf-8',index_col=0,header=None,
                         names=["label", "textA", "textB"])
train_data.drop_duplicates(inplace=True)
df = pd.DataFrame(train_data)
grouped = df.groupby(['label']).head(3000)
grouped = grouped.sort_values(['label'], ascending=True)
#
print(grouped)
grouped.to_csv('E:\\tmp\\2\\tagging2.tsv',
                  sep='\t',
                  encoding='utf-8',
                  header=None)