Question

我正在运行此代码，并得到我无法解决的错误：

import xgboost as xgb1
from sklearn.grid_search import GridSearchCV

cv_params = {'max_depth': [3,5,7], 'min_child_weight': [1,3,5] }

ind_params = {
    'learning_rate': 0.1,
    'n_estimators': 1000,
    'seed': 0,
    'subsample': 0.8,
    'colsample_bytree': 0.8,
    'objective': 'count:poisson'
    }

optimized_GBM = GridSearchCV(xgb1.XGBRegressor(**ind_params), 
                             cv_params, 
                             scoring = 'mean_squared_error',
                             cv = 5, 
                             n_jobs = -1) 

optimized_GBM.fit(train_X, train_y)

print(optimized_GBM.best_score_)
print(optimized_GBM.best_params_)

错误：

validation.py in _assert_all_finite(X=array([nan, nan, nan, ..., nan, nan, nan], dtype=float32))
     39     # everything is finite; fall back to O(n) space np.isfinite to prevent
     40     # false positives from overflow in sum method.
     41     if (X.dtype.char in np.typecodes['AllFloat'] and not np.isfinite(X.sum())
     42             and not np.isfinite(X).all()):
     43         raise ValueError("Input contains NaN, infinity"
---> 44                          " or a value too large for %r." % X.dtype)
        X.dtype = dtype('float32')
     45 
     46 
     47 def assert_all_finite(X):
     48     """Throw a ValueError if X contains NaN or infinity.

ValueError: Input contains NaN, infinity or a value too large for dtype('float32').

所有X个特征的类型均为float64，而y是整数。我还尝试将X更改为float32。我检查了是否没有NaN和inf数据：

XY = XY.replace([np.inf, -np.inf], np.nan).dropna(subset=XY.columns, how="all").round(2)
scaler = preprocessing.MinMaxScaler()
scaler.fit(XY.drop("DEPARTURE_DELAY_MIN",axis=1))
standardized_df = pd.DataFrame(scaler.transform(XY.drop("DEPARTURE_DELAY_MIN",axis=1)),columns=XY.drop("DEPARTURE_DELAY_MIN",axis=1).columns)

X = standardized_df
y = XY.DEPARTURE_DELAY_MIN.astype("int")

X = X.astype("double")
y = y.astype("int")

# Split the data set into train and test sets
train_X, test_X, train_y, test_y = train_test_split(X, y, test_size=0.25)

我不明白会发生什么。

XGBoost：由于ValueError而无法拟合模型

0 个答案: