all_data = list(zip(*data))
lengths = sorted(list(set([len(x[bucket_ind]) for x in all_data])))
if n_bucket > len(lengths):
n_bucket = len(lengths)

splits = np.array(lengths)[
    (np.linspace(0, 1, 5, endpoint=False) * len(lengths)).astype(int)
].tolist()

splits += [np.inf] #np.inf无限大的正整数

if debug:
    print(splits)

ind_data = {}
for x in all_data:
    l = len(x[bucket_ind])
    for ind, s in enumerate(splits[:-1]):
        if l >= s and l <= splits[ind + 1]:
            if ind not in ind_data:
                ind_data[ind] = []
            ind_data[ind].append(x)
            break

inds = sorted(list(ind_data.keys()))
ind_p = [len(ind_data[x]) / len(all_data) for x in inds]
if debug:
    print(np.sum(ind_p), ind_p)