Question

我有一个包含12000行和160列的数据集，所有数据都是整数。当我想使用此代码查找偏差方差折衷时，我收到内存错误：

for i, step in enumerate(steps):
    # Generate polynomial features
    poly = PolynomialFeatures(step, interaction_only=True)
    pX_train = poly.fit_transform(X_train_new_subset.astype(float))
    pX_test = poly.transform(X_test_new_subset.astype(float))

    # Create model instance
    model = KNeighborsRegressor(n_neighbors=5)

    # Train model
    model.fit(pX_train, y_train_new_subset.astype(float))

    # Make Predictions
    y_train_pred = model.predict(pX_train)
    y_test_pred = model.predict(pX_test)

    # Calculate Error and accuracy
    tr_score[i]=r2_score(y_train_new_subset.astype(float), y_train_pred)
    tst_score[i]=r2_score(y_test_new_subset.astype(float), y_test_pred)
    trainErr[i] = mean_squared_error(y_train_new_subset.astype(float), y_train_pred)
    testErr[i] = mean_squared_error(y_test_new_subset.astype(float), y_test_pred)

此代码有什么问题吗？

---------------------------------------------------------------------------
MemoryError                               Traceback (most recent call last)
<ipython-input-45-1c32333265a0> in <module>
     10     # Generate polynomial features
     11     poly = PolynomialFeatures(step, interaction_only=True)
---> 12     pX_train = poly.fit_transform(X_train_new_subset)
     13     pX_test = poly.transform(X_test_new_subset)
     14 

d:\Users\rmardani\Anaconda3\lib\site-packages\sklearn\base.py in fit_transform(self, X, y, **fit_params)
    460         if y is None:
    461             # fit method of arity 1 (unsupervised transformation)
--> 462             return self.fit(X, **fit_params).transform(X)
    463         else:
    464             # fit method of arity 2 (supervised transformation)

d:\Users\rmardani\Anaconda3\lib\site-packages\sklearn\preprocessing\data.py in transform(self, X)
   1484             XP = sparse.hstack(columns, dtype=X.dtype).tocsc()
   1485         else:
-> 1486             XP = np.empty((n_samples, self.n_output_features_), dtype=X.dtype)
   1487             for i, comb in enumerate(combinations):
   1488                 XP[:, i] = X[:, comb].prod(1)

MemoryError:

偏差方差折衷的内存错误python

0 个答案: