CODEWITHSUNDEEP
pythonarrayspandasscikit-learncross-validation
Sahil
Sahil

Reputation: 439

TypeError: only length-1 arrays can be converted to Python scalars, while using Kfold cross Validation

I am trying to use Kfold cross valiadtion for my model, but get this error while doing so. I know that KFold only accepts 1D arrays but even after converting the length input to an array its giving me this problem.

from sklearn.ensemble import ExtraTreesClassifier, RandomForestClassifier
from sklearn.cross_validation import train_test_split
from sklearn.cross_validation import KFold

if __name__ == "__main__":
  np.random.seed(1335)

  verbose = True
  shuffle = False
  n_folds = 5

  y = np.array(y)

  if shuffle: 
    idx = np.random.permutation(y.size)
    X_train = X_train[idx]
    y = y[idx]

  skf = KFold(y, n_folds) 

  models = [RandomForestClassifier(n_estimators=100, n_jobs=-1, criterion='gini'),ExtraTreesClassifier(n_estimators=100, n_jobs=-1, criterion='entropy')]

  print("Stacking in progress")
  A = []
  for j, clf in enumerate(models):
     print(j, clf)

     for i, (itrain, itest) in enumerate(skf):
        print("Fold :", i)
        x_train = X_train[itrain]
        x_test = X_train[itest]
        y_train = y[itrain]
        y_test = y[itest]
        print(x_train.shape, x_test.shape)

        print(len(x_train), len(x_test))

        clf.fit(x_train, y_train)
        pred = clf.predict_proba(x_test)
        A.append(pred)

I get the error for the line "skf = KFold(y, n_folds)". Any help with this will be appreciated.

Upvotes: 0

Views: 453

Answers (1)

P. Camilleri
P. Camilleri

Reputation: 13218

From its doc, KFold() does not expect y as an input, but only the number of splits (n_folds).

Once you have an instance of KFold, you do myKfold.split(x) (x being all of your input data) to obtain an iterator yielding train and test indices. Example copy pasted from sklearn doc:

>>> from sklearn.model_selection import KFold
>>> X = np.array([[1, 2], [3, 4], [1, 2], [3, 4]])
>>> y = np.array([1, 2, 3, 4])
>>> kf = KFold(n_splits=2)
>>> kf.get_n_splits(X)
2
>>> print(kf)  
KFold(n_splits=2, random_state=None, shuffle=False)
>>> for train_index, test_index in kf.split(X):
...    print("TRAIN:", train_index, "TEST:", test_index)
...    X_train, X_test = X[train_index], X[test_index]
...    y_train, y_test = y[train_index], y[test_index]
TRAIN: [2 3] TEST: [0 1]
TRAIN: [0 1] TEST: [2 3]

Upvotes: 1

Related Questions