I’m testing XGBoost with the following code (data loaded already):
import xgboost
from xgboost import XGBClassifier
model=XGBClassifier(n_estimators=1400,max_depth= 7,
learning_rate=0.001, subsample=0.1)
model.fit(X_train, y_train)
The second line of code model.fit(X_train, y_train) reports the following ERROR messages:
File ~/.local/lib/python3.8/site-packages/xgboost/core.py:506, in _deprecate_positional_args.<locals>.inner_f(*args, **kwargs)
504 for k, arg in zip(sig.parameters, args):
505 kwargs[k] = arg
--> 506 return f(**kwargs)
File ~/.local/lib/python3.8/site-packages/xgboost/sklearn.py:1181, in XGBClassifier.fit(self, X, y, sample_weight, base_margin, eval_set, eval_metric, early_stopping_rounds, verbose, xgb_model, sample_weight_eval_set, base_margin_eval_set, feature_weights, callbacks)
1176 if (
1177 self.classes_.shape != expected_classes.shape
1178 or not (self.classes_ == expected_classes).all()
1179 ):
1180 raise ValueError(label_encoding_check_error)
-> 1181 elif _is_cupy_array(y):
1182 import cupy as cp # pylint: disable=E0401
1184 self.classes_ = cp.unique(y)
File ~/.local/lib/python3.8/site-packages/xgboost/data.py:596, in _is_cupy_array(data)
594 except ImportError:
595 return False
--> 596 return isinstance(data, cupy.ndarray)
AttributeError: module 'cupy' has no attribute 'ndarray'
It looks to have some relationship with XGBoost, as well as copy’s function _is_cupy_array. Any suggestions please?