0

我一直在尝试使用稀疏矩阵进行关联规则挖掘，但它一直在抛出AttributeError: dtypes not found. 我知道错误的含义，但我无法解决它。这就是我所拥有的：

import pandas as pd
import scipy
from mlxtend.frequent_patterns import fpgrowth
from mlxtend.frequent_patterns import apriori, association_rules

test_data = pd.read_csv('.../test_data.csv')

test_data = pd.DataFrame(test_data, dtype = bool)
sparse_df = scipy.sparse.csr_matrix(test_data.values)
sparse_df

结果：

 <1000x10 sparse matrix of type '<class 'numpy.bool_'>'
    with 1257 stored elements in Compressed Sparse Row format>

直到这里一切都很好并且工作正常，但是当我运行下一个块时，它会弹出错误。

apriori(sparse_df, min_support = 0.3, use_colnames = True, verbose = 1)

我已经在其他问题中检查了该问题的其他“解决方案”，但没有一个对我有用。

整个错误

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-259-ac3b4693270b> in <module>
----> 1 apriori(sparse_df, min_support = 0.3, use_colnames = True, verbose = 1)

~\anaconda3\lib\site-packages\mlxtend\frequent_patterns\apriori.py in apriori(df, min_support, use_colnames, max_len, verbose, low_memory)
    236                          'Got %s.' % min_support)
    237 
--> 238     fpc.valid_input_check(df)
    239 
    240     if hasattr(df, "sparse"):

~\anaconda3\lib\site-packages\mlxtend\frequent_patterns\fpcommon.py in valid_input_check(df)
     98 
     99     # Fast path: if all columns are boolean, there is nothing to checks
--> 100     all_bools = df.dtypes.apply(pd.api.types.is_bool_dtype).all()
    101     if not all_bools:
    102         # Pandas is much slower than numpy, so use np.where on Numpy arrays

~\anaconda3\lib\site-packages\scipy\sparse\base.py in __getattr__(self, attr)
    685             return self.getnnz()
    686         else:
--> 687             raise AttributeError(attr + " not found")
    688 
    689     def transpose(self, axes=None, copy=False):

AttributeError: dtypes not found
4

0 回答 0