discretizer.py 文件源码

def _discretize_by_gmm(col, num_bins, random_state):
    nan_idx = col[col.isnull()].index
    gmm = GaussianMixture(n_components=num_bins,covariance_type='full',random_state=random_state)
    gmm = gmm.fit(X=np.expand_dims(col.dropna(), 1))
    if col.isnull().sum() == 0:
        group = gmm.predict(X=np.expand_dims(col, 1))
    else:
        group = gmm.predict(X=np.expand_dims(col.dropna(), 1)).astype(float)
        for idx in nan_idx:
            group = np.insert(group,idx,np.nan)
    return pd.Series(group)