Pandas拼接concat使用方法

df1 = pd.DataFrame（{ 
    'name'：['A'，'B'，'C'，'D']，
    'math'：[60,89,82,70]，
    'physics'：[66， 95,83,66]，
    'chemistry'：[61,91,77,70] 
}）
df2 = pd.DataFrame（{ 
    'name'：['E'，'F'，'G'，'H']，
    'math'：[66,95,83,66]，
    'physics'：[60， 89,82,70]，
    'chemistry'：[90,81,78,90] 
}）

pd.concat（[df1，df2]）

pd.concat（[df1，df2]，ignore_index = True）

pd.concat（[df1，df2]，axis = 1）

try:
    pd.concat([df1,df2], verify_integrity=True)
except ValueError as e:
    print('ValueError', e)
ValueError: Indexes have overlapping values: Int64Index([0, 1, 2, 3], dtype='int64')

res = pd.concat（[df1，df2]，keys = ['Year 1'，'Year 2']）
res

res.loc['Year 1']

pd.concat(
    [df1，df2]，
    keys = ['Year 1'，'Year 2']，
    names = ['Class'，None]，
)

pd.concat(
    [df1, df2], 
    keys=['Year 1', 'Year 2'],
    names=['Class', None],
).reset_index(level=0)   
# reset_index(level='Class')

pd.concat([df1, df2], sort=True)

custom_sort = ['math', 'chemistry', 'physics', 'name']
res = pd.concat([df1, df2])
res[custom_sort]

import pathlib2 as pl2
ps = pl2.Path('data/sp3')
res = None
for p in ps.glob('*.csv'):
    if res is None:
        res = pd.read_csv(p)
    else:
        res = pd.concat([res, pd.read_csv(p)])

import pathlib2 as pl2
ps = pl2.Path('data/sp3')
dfs = (
    pd.read_csv(p, encoding='utf8') for p in ps.glob('*.csv')
)
res = pd.concat(dfs)
res

# for-loop solution
298 ms ± 11.8 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
# list comprehension solution
153 ms ± 6 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)