pandas 的非循环替代方案

高效编写 Python 代码

Logan Thomas

Scientific Software Technical Trainer, Enthought

print(baseball_df.head())

  Team League  Year   RS   RA   W    G  Playoffs
0  ARI     NL  2012  734  688  81  162         0
1  ATL     NL  2012  700  600  94  162         1
2  BAL     AL  2012  712  705  93  162         1
3  BOS     AL  2012  734  806  69  162         0
4  CHC     NL  2012  613  759  61  162         0

def calc_run_diff(runs_scored, runs_allowed):

    run_diff = runs_scored - runs_allowed

    return run_diff
高效编写 Python 代码

用循环计算分差

run_diffs_iterrows = []

for i,row in baseball_df.iterrows():
    run_diff = calc_run_diff(row['RS'], row['RA'])
    run_diffs_iterrows.append(run_diff)

baseball_df['RD'] = run_diffs_iterrows
print(baseball_df)

     Team League  Year   RS   RA    W    G  Playoffs   RD
0     ARI     NL  2012  734  688   81  162         0   46
1     ATL     NL  2012  700  600   94  162         1  100
2     BAL     AL  2012  712  705   93  162         1    7
...
高效编写 Python 代码

pandas 的 .apply() 方法

  • 接收一个函数并作用于 DataFrame
    • 需指定应用轴（列为 0；行为 1）
  • 可配合匿名函数（lambda）使用
  • 示例：
baseball_df.apply(

    lambda row: calc_run_diff(row['RS'], row['RA']),

    axis=1
)
高效编写 Python 代码

用 .apply() 计算分差

run_diffs_apply = baseball_df.apply(
         lambda row: calc_run_diff(row['RS'], row['RA']),
         axis=1)

baseball_df['RD'] = run_diffs_apply
print(baseball_df)

     Team League  Year   RS   RA    W    G  Playoffs   RD
0     ARI     NL  2012  734  688   81  162         0   46
1     ATL     NL  2012  700  600   94  162         1  100
2     BAL     AL  2012  712  705   93  162         1    7
...
高效编写 Python 代码

方法对比

%%timeit
run_diffs_iterrows = []

for i,row in baseball_df.iterrows():
    run_diff = calc_run_diff(row['RS'], row['RA'])
    run_diffs_iterrows.append(run_diff)

baseball_df['RD'] = run_diffs_iterrows

每个循环 86.8 ms ± 3 ms（7 次运行的均值 ± 标准差，每次 10 循环）
高效编写 Python 代码

方法对比

%%timeit
run_diffs_apply = baseball_df.apply(
         lambda row: calc_run_diff(row['RS'], row['RA']),
         axis=1)

baseball_df['RD'] = run_diffs_apply

每个循环 30.1 ms ± 1.75 ms（7 次运行的均值 ± 标准差，每次 10 循环）
高效编写 Python 代码

练习使用 pandas 的 .apply() 方法！

高效编写 Python 代码

Preparing Video For Download...