我在 pandas dataframe 列中有字典作为值。 我想将键列和值作为列值

Question

So i'm working on a dataframe which has a key-value pair as its value in columns. Is there a way to make the keys as column name while only keeping the value left in the column.

Currently i have something like this:

    >0                          1                       2   
    >{'1536235175000': 26307.9} {'1536235176000': 0}    {'1536236701000': 2630} 
    >{'1536239919000': 1028127} {'1536239921000': 0}    NaN 
    >{'1536242709000': 2629.6}  {'1536242711000': 0}    NaN 

Answer 1

If you want to keep the row index, you can agg every row as as list and explode them.

obj = df.apply(lambda x: list(x), axis=1).explode().dropna()
dfn = pd.DataFrame(obj.tolist(), index=obj.index)
dfn.stack().unstack()

#        1536235175000  1536235176000  1536236701000  1536239919000  \
# 0            26307.9            0.0         2630.0            NaN   
# 1                NaN            NaN            NaN      1028127.0   
# 2                NaN            NaN            NaN            NaN   

#        1536239921000  1536242709000  1536242711000  
# 0                NaN            NaN            NaN  
# 1                0.0            NaN            NaN  
# 2                NaN         2629.6            0.0

Answer 2

Check with concat

pd.concat([pd.Series(df[x].tolist()) for x in df.columns], keys=df.columns, axis=1)

Answer 3

If lakes is your DataFrame, you can do something like

area_dict = dict(zip(lakes.area, lakes.count) )