Python 在连续数据上计算DCG和NDCG

是否可以在python中对连续数据计算NDCG？目前，我正试图计算NDCG的数据是连续的性质，但得到一个例外。我如何解决这个问题

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-96-0eec5bfa270d> in <module>
     38 #main_extract_research_work()
     39 #calculate_research_score(0.5,0.5)
---> 40 calculate_final_scores()

<ipython-input-95-ba2bcdacd296> in calculate_final_scores()
     32         relevance_score = np.array(order)
     33         true_relevance = np.sort(relevance_score)[::-1]
---> 34         score = m.ndcg_score(true_relevance, relevance_score)

~\Anaconda3\lib\site-packages\sklearn\metrics\_ranking.py in ndcg_score(y_true, y_score, k, sample_weight, ignore_ties)
   1417     y_score = check_array(y_score, ensure_2d=False)
   1418     check_consistent_length(y_true, y_score, sample_weight)
-> 1419     _check_dcg_target_type(y_true)
   1420     gain = _ndcg_sample_scores(y_true, y_score, k=k, ignore_ties=ignore_ties)
   1421     return np.average(gain, weights=sample_weight)

~\Anaconda3\lib\site-packages\sklearn\metrics\_ranking.py in _check_dcg_target_type(y_true)
   1161         raise ValueError(
   1162             "Only {} formats are supported. Got {} instead".format(
-> 1163                 supported_fmt, y_type))
   1164 
   1165 

ValueError: Only ('multilabel-indicator', 'continuous-multioutput', 'multiclass-multioutput') formats are supported. Got continuous instead

from sklearn import metrics as m

for i in range(len(size)):
        order=[education_scores[i],experience_scores[i],technical_skills_scores[i],soft_skills_scores[i]]
        relevance_score = np.array(order)
        true_relevance = np.sort(relevance_score)[::-1]
        score = m.ndcg_score(true_relevance, relevance_score)