当MultiIndex构造函数不可用时，很难提供帮助.您可以使用stack和一些索引方法reshape 数据帧:
>>> (df.set_index(df.columns[:4].tolist()) # Market, Product, City, Territory
.rename_axis(index=df.columns[:4].droplevel(0), # Flat them
columns=['Date', None]) # Define column names
.stack('Date', sort=False).reset_index()) # Reshape your dataframe
Market Product City Territory Date VALUES Values MARKET SHARE VALUES GROWTH VALUES GEO. SHARE UNITS UNITS MARKET SHARE UNITS GROWTH UNITS GEO. SHARE
0 0 0 0 0 23-Jan 1 1 1 1 1 1 1 1
1 0 0 0 0 23-Feb 2 2 2 2 2 2 2 2
最小工作示例:
data = {('', 'Market'): {0: 0},
('', 'Product'): {0: 0},
('', 'City'): {0: 0},
('', 'Territory'): {0: 0},
('23-Jan', 'VALUES'): {0: 1},
('23-Jan', 'Values MARKET SHARE'): {0: 1},
('23-Jan', 'VALUES GROWTH'): {0: 1},
('23-Jan', 'VALUES GEO. SHARE'): {0: 1},
('23-Jan', 'UNITS'): {0: 1},
('23-Jan', 'UNITS MARKET SHARE'): {0: 1},
('23-Jan', 'UNITS GROWTH'): {0: 1},
('23-Jan', 'UNITS GEO. SHARE'): {0: 1},
('23-Feb', 'VALUES'): {0: 2},
('23-Feb', 'Values MARKET SHARE'): {0: 2},
('23-Feb', 'VALUES GROWTH'): {0: 2},
('23-Feb', 'VALUES GEO. SHARE'): {0: 2},
('23-Feb', 'UNITS'): {0: 2},
('23-Feb', 'UNITS MARKET SHARE'): {0: 2},
('23-Feb', 'UNITS GROWTH'): {0: 2},
('23-Feb', 'UNITS GEO. SHARE'): {0: 2}}
df = pd.DataFrame(data)
print(df)
# Output
23-Jan ... 23-Feb
Market Product City Territory VALUES Values MARKET SHARE ... VALUES GROWTH VALUES GEO. SHARE UNITS UNITS MARKET SHARE UNITS GROWTH UNITS GEO. SHARE
0 0 0 0 0 1 1 ... 2 2 2 2 2 2
[1 rows x 20 columns]
I want to create a Python function that transforms this DataFrame into the following format:
How can I achieve this transformation using Python and Pandas?