python - renaming column labels in multiindex df (built from excel) -
i'm reading excel file , manipulating content pandas df.
heres excel content looks (fig 1). , here's want df (fig 2) 1 exception - want rename column labels @ index level 0, "a4+" "all ind" , "a18-49" "comm demo".
the problem have when rename column labels come out inverted, not sure why?
here's code use..
full_path_to_excel_file = path_runs+excel_file_name+'.xls' df = pd.read_excel(full_path_to_excel_file, 0, header=none, index_col=0) df = df.iloc[11:,:] df = df.fillna(method='ffill', axis=1) df.columns=pd.multiindex.from_arrays(df[:2].values) #just name of index df.index.name=none #remove 3 rows used column names df = df[pd.notnull(df.index)] df = df.drop(['channel group ...']) df.columns = df.columns.droplevel() df.columns.set_levels(['all ind', 'comm demo'], 0, inplace=true)
the excel used here: https://www.dropbox.com/s/vvpz8p7kzpcph6q/time%20kpi%20000s%20bg.xls?dl=0
fig1
fig 2
use set_levels
consider df
df = pd.dataframe(np.arange(8).reshape(-1, 4), columns=pd.multiindex.from_product([['a4+', 'a18-49'], list('xy')])) df
m = {'a4+': 'all ind', 'a18-49': 'comm demo'} df.columns.set_levels(df.columns.levels[0].to_series().map(m), level=0, inplace=true) df
Comments
Post a Comment