python - 使用 Python Pandas 进行数据分析

Question

我是 Pandas 库的新手，需要一些帮助。我有两列这样的:

Test Result       Risk Rating
  Fail               Low                   
  Pass               Medium
  Skip               High
  Pass               Low                   
  Fail               Medium
  Pass               High
  Skip               Low                   
  Fail               Medium
  Fail               High

现在，我需要从“测试结果”列中找到失败、通过和跳过的总数，并且我能够做到这一点。但是，我还需要从测试结果列中获取“失败”的总数，并从风险评级列中获取“低”的总数。同样，“中等失败”总数等等。我的最终结果应该是这样的:

Fail (Low Risk Rating) = 1
Fail (Medium Risk Rating) = 2
Fail (High Risk Rating) = 1
Pass (Low Risk Rating) = 1
Pass (Medium Risk Rating) = 1
Pass (High Risk Rating) = 1
Skip (Low Risk Rating) = 1
Skip (Medium Risk Rating) = 0
Skip (High Risk Rating) = 1

我该怎么做？任何帮助将不胜感激。

Answer 1

我认为你需要groupby按列和聚合 size :

df = df.groupby(['Test Result', 'Risk Rating']).size().reset_index(name='counts')
print (df)
  Test Result Risk Rating  counts
0        Fail        High       1
1        Fail         Low       1
2        Fail      Medium       2
3        Pass        High       1
4        Pass         Low       1
5        Pass      Medium       1
6        Skip        High       1
7        Skip         Low       1

也许更好的是数据透视表 unstack :

df = df.groupby(['Test Result', 'Risk Rating']).size().unstack(fill_value=0)
print (df)
Risk Rating  High  Low  Medium
Test Result                   
Fail            1    1       2
Pass            1    1       1
Skip            1    1       0

或更慢的解决方案 crosstab :

df = pd.crosstab(df['Test Result'], df['Risk Rating'])
print (df)
Risk Rating  High  Low  Medium
Test Result                   
Fail            1    1       2
Pass            1    1       1
Skip            1    1       0

如果需要 0 缺失值，请添加 stack :

df = df.groupby(['Test Result', 'Risk Rating'])
       .size()
       .unstack(fill_value=0)
       .stack()
       .reset_index(name='counts')
print (df)
  Test Result Risk Rating  counts
0        Fail        High       1
1        Fail         Low       1
2        Fail      Medium       2
3        Pass        High       1
4        Pass         Low       1
5        Pass      Medium       1
6        Skip        High       1
7        Skip         Low       1
8        Skip      Medium       0