shap.Cohorts
- class shap.Cohorts(**kwargs: Explanation)
- 一个 - Explanation对象的集合,通常每个对象解释相似样本的集群。- 示例 - Cohorts对象可以通过多种方式初始化。- 通过显式指定 cohorts - >>> exp = Explanation( ... values=np.random.uniform(low=-1, high=1, size=(500, 5)), ... data=np.random.normal(loc=1, scale=3, size=(500, 5)), ... feature_names=list("abcde"), ... ) >>> cohorts = Cohorts( ... col_a_neg=exp[exp[:, "a"].data < 0], ... col_a_pos=exp[exp[:, "a"].data >= 0], ... ) >>> cohorts <shap._explanation.Cohorts object with 2 cohorts of sizes: [(198, 5), (302, 5)]> - 或使用 - Explanation.cohorts()方法- >>> cohorts2 = exp.cohorts(3) >>> cohorts2 <shap._explanation.Cohorts object with 3 cohorts of sizes: [(182, 5), (12, 5), (306, 5)]> - Explanation接口的大部分也在- Cohorts中公开。例如,要检索跨所有 cohorts 的列 ‘a’ 对应的 SHAP 值,您可以使用- >>> cohorts[..., 'a'].values <shap._explanation.Cohorts object with 2 cohorts of sizes: [(198,), (302,)]> - 要实际检索特定 - Explanation的值,您需要通过- Cohorts.cohorts()属性访问它- >>> cohorts.cohorts["col_a_neg"][..., 'a'].values array([...]) # truncated - __init__(**kwargs: Explanation) None
 - 方法 - __init__(**kwargs)- 属性 - cohorts 的内部集合,存储为字典。 - property cohorts: dict[str, Explanation]
- cohorts 的内部集合,存储为字典。