MisterLee's recent timeline updates MisterLee's recent timeline updates |
MisterLee2006年美国《时代周刊》年度风云人物 V2EX member #472680, joined on 2020-02-27 11:38:36 +08:00 |
MisterLee's recent replies
Mar 19, 2020 Replied to a topic by thinszx › Python › pandas 关于 groupby 的分组保存问题：如何将分组完以后的值按照某个列分别存为新 dataframe？ |
df_1 = df.groupby('A').apply(lambda x: x.sort_values('B')).reset_index(drop = True)
for item in df_1['A'].unique():
df_2 = df_1[df_1['A'] == item]
df_2.to_xxx #导出
for item in df_1['A'].unique():
df_2 = df_1[df_1['A'] == item]
df_2.to_xxx #导出
Feb 27, 2020 Replied to a topic by qazwsxkevin › Python › 有两个这样的字典列表，如何按字典的 key 值进行去重合并？ 比较高效率的做法？ |
——Python&Pandas
a_df= pd.DataFrame(a_list)
b_df= pd.DataFrame(b_list)
a_b_df = pd.concat([a_df, b_df], axis=0).drop_duplicates(['型号', '重量', '审核人'], keep='first')
c_list = []
for _, item in a_b_df.iterrows():
c_list.append(item.dropna().to_dict())
a_df= pd.DataFrame(a_list)
b_df= pd.DataFrame(b_list)
a_b_df = pd.concat([a_df, b_df], axis=0).drop_duplicates(['型号', '重量', '审核人'], keep='first')
c_list = []
for _, item in a_b_df.iterrows():
c_list.append(item.dropna().to_dict())
Select Language