# Skopt functions
from skopt import BayesSearchCV
from skopt import gp_minimize # Bayesian optimization using Gaussian Processes
from skopt.space import Real, Categorical, Integer
from skopt.utils import use_named_args # decorator to convert a list of parameters to named arguments
from skopt.callbacks import DeadlineStopper # Stop the optimization before running out of a fixed budget of time.
from skopt.callbacks import VerboseCallback # Callback to control the verbosity
# Stop the optimization If the last two positions at which the objective has been evaluated are less than delta
from skopt.callbacks import DeltaXStopper
from gen_feas import load_data
# Uploading the Boston dataset
# X, y = load_boston(return_X_y=True)
train, test, no_features, features = load_data()
X = train[features].values
y = train['target'].astype('int32')
# Transforming the problem into a classification (unbalanced)
# y_bin = (y > np.percentile(y, 90)).astype(int)
clf = lgb.LGBMClassifier(boosting_type='gbdt',
class_weight='balanced',
objective='binary',
n_jobs=-1,
verbose=0)
search_spaces = {
'learning_rate': Real(0.01, 1.0, 'log-uniform'),
'num_leaves': Integer(2, 500),
'max_depth': Integer(0, 500),
'min_child_samples': Integer(0, 200),
'max_bin': Integer(100, 100000),
xgb_params = {
'objective': 'binary:logistic',
'max_depth': 5,
'n_estimators': 100000,
'learning_rate': 0.1,
'nthread': 4,
'subsample': 0.7,
'colsample_bytree': 0.7,
'min_child_weight': 3,
'n_jobs':-1
}
n_fold = 5
num_classes = 2
print("分类个数num_classes:{}".format(num_classes))
folds = StratifiedKFold(n_splits=n_fold, random_state=1314)
train, y, test, features = load_data()
X = train[features]
print(y.value_counts())
X_test = test[features]
result_dict_lgb = train_model_classification(X=X,
X_test=X_test,
y=y,
params=xgb_params,
num_classes=num_classes,
folds=folds,
model_type='xgb',
eval_metric='logloss',
plot_feature_importance=False,
verbose=10,
'min_child_weight': 3
}
cat_paras = {
'iterations': 50,
'learning_rate': 0.1,
'od_type': 'Iter',
'l2_leaf_reg': 3,
'depth': 10
}
n_fold = 5
nums = [3, 4, 5, 11]
for num_classes in nums:
print("分类个数num_classes:{}".format(num_classes))
n_fold = 5
folds = StratifiedKFold(n_splits=n_fold, random_state=1314)
train, _, test, features = load_data(num_classes)
X = train[features]
X['Year-Of-Publication'] = X['Year-Of-Publication'].astype(int)
X_test = test[features]
X_test['Year-Of-Publication'] = X_test['Year-Of-Publication'].astype(int)
y = train['Book-Rating'].astype(int)
print(y.value_counts())
X = train[features]
X_test = test[features]
result_dict_lgb = train_model_classification(X=X,
X_test=X_test,
y=y,
params=xgb_params,
num_classes=num_classes,