analyze_metaclusters.py 文件源码

def _compute_dispersion_matrix(X, labels):
    n = len(np.unique(labels))
    dist = np.zeros((n, n))
    ITR = list(itertools.combinations_with_replacement(range(n), 2))
    for i, j in tqdm(ITR):

        if i == j:
            d = pdist(X[labels == i], metric='cosine')
        else:
            d = cdist(X[labels == i], X[labels == j], metric='cosine')
            # Only take upper diagonal (+diagonal elements)
            d = d[np.triu_indices(n=d.shape[0], m=d.shape[1], k=0)]

        dist[i, j] = dist[j, i] = d.mean()

    return dist