Блог

Я пытаюсь построить модель FRNN:

 for i, owa_weights, k in [(1, strict(), 1), (2, additive(), 20)]:
  # Create an instance of the FRNN classifier, construct the model, and query on the test set.
  clf = FRNN(nn_search=nn_search, upper_weights=owa_weights, lower_weights=owa_weights, lower_k=k, upper_k=k)
  cl = clf.construct(X_train, y)
  scores = model(X_test)
 

Но я получаю ошибку:

 ---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-81-4d508eeca158> in <module>()
      7   # Create an instance of the FRNN classifier, construct the model, and query on the test set.
      8   clf = FRNN(nn_search=nn_search, upper_weights=owa_weights, lower_weights=owa_weights, lower_k=k, upper_k=k)
----> 9   cl = clf.construct(X_train, y)
     10   scores = model(X_test)

2 frames
/content/frlearn/neighbours/classifiers.py in <listcomp>(.0)
     24     def construct(self, X, y) -> Model:
     25         model: FuzzyRoughEnsemble.Model = super().construct(X, y)
---> 26         Cs = [X[np.where(y == c)] for c in model.classes]
     27         model.upper_approximations = self.upper_approximator and [self.upper_approximator.construct(C) for C in Cs]
     28         co_Cs = [X[np.where(y != c)] for c in model.classes]

IndexError: index 73922 is out of bounds for axis 0 with size 73920
 

Форма X_train равна (73920 300), а затем я пытаюсь запустить вот так:

 cl = clf.construct(X_train[-2:73920], y)
 

Но я получаю ошибку:

 IndexError: index 49 is out of bounds for axis 0 with size 2
 

Вот подробная информация о сообщениях об ошибках: