python dataframe數值替換

需求：將csv檔案中的值為-8888的因子值替換為當日全市場因子的中值

原始資料：

解析：實現：

import pandas as pd

def filter2median(invalid_number):
with open(r'c:\users\administrator\documents\wechat files\xzr8510776\filestorage\file\2019-04\test.csv',
encoding="utf-8") as f:
df = pd.read_csv(f)
feature_list = df.columns.tolist()[2:]
median = df.groupby(df['date']).median()
# for index in df.index:
#     date = df.loc[index, 'date']
#     for feature in feature_list:
#         if df.loc[index, feature] == invalid_number:
#             df.loc[index, feature] = media.loc[date][feature]
for feature in feature_list:
df.loc[df[feature] == invalid_number, feature] = median.loc[
df.loc[df[feature] == invalid_number, 'date'], feature].tolist()
print(df)
# df.to_csv("r.csv")
if __name__ == '__main__':
filter2median(-8888)

python DataFrame合併方法

python 合併dataframe方法 dataframe.merge left,right,how inner on none,left on none,right on none,left index false,right index false,sort false,suffixes x ...

Python Dataframe 兩列相除

import pandas as pd df pd.read csv 離線資料.csv 新建乙個dataframe df data pd.dataframe columns 總數 人數 平均值 df data 總數 人數 df df data 平均值 df data.lambda x x 總數 x ...

python DataFrame匯出到檔案

outputpath d users chen lib desktop fenci.csv df.to csv outputpath,sep index false,header false 2.dataframe匯出到xlsx檔案 outputpath d users chen lib deskt...