# coding: utf-8
"""Library with training routines of LightGBM."""
import collections
import copy
from operator import attrgetter
from pathlib import Path
from typing import Any, Callable, Dict, List, Optional, Tuple, Union
import numpy as np
from . import callback
from .basic import Booster, Dataset, LightGBMError, _ConfigAliases, _InnerPredictor, _log_warning
from .compat import SKLEARN_INSTALLED, _LGBMGroupKFold, _LGBMStratifiedKFold
_LGBM_CustomObjectiveFunction = Callable[
[Union[List, np.ndarray], Dataset],
Tuple[Union[List, np.ndarray], Union[List, np.ndarray]]
]
_LGBM_CustomMetricFunction = Callable[
[Union[List, np.ndarray], Dataset],
Tuple[str, float, bool]
]
[docs]def train(
params: Dict[str, Any],
train_set: Dataset,
num_boost_round: int = 100,
valid_sets: Optional[List[Dataset]] = None,
valid_names: Optional[List[str]] = None,
fobj: Optional[_LGBM_CustomObjectiveFunction] = None,
feval: Optional[Union[_LGBM_CustomMetricFunction, List[_LGBM_CustomMetricFunction]]] = None,
init_model: Optional[Union[str, Path, Booster]] = None,
feature_name: Union[List[str], str] = 'auto',
categorical_feature: Union[List[str], List[int], str] = 'auto',
early_stopping_rounds: Optional[int] = None,
evals_result: Optional[Dict[str, Any]] = None,
verbose_eval: Union[bool, int, str] = 'warn',
learning_rates: Optional[Union[List[float], Callable[[int], float]]] = None,
keep_training_booster: bool = False,
callbacks: Optional[List[Callable]] = None
) -> Booster:
"""Perform the training with given parameters.
Parameters
----------
params : dict
Parameters for training.
train_set : Dataset
Data to be trained on.
num_boost_round : int, optional (default=100)
Number of boosting iterations.
valid_sets : list of Dataset, or None, optional (default=None)
List of data to be evaluated on during training.
valid_names : list of str, or None, optional (default=None)
Names of ``valid_sets``.
fobj : callable or None, optional (default=None)
Customized objective function.
Should accept two parameters: preds, train_data,
and return (grad, hess).
preds : list or numpy 1-D array
The predicted values.
Predicted values are returned before any transformation,
e.g. they are raw margin instead of probability of positive class for binary task.
train_data : Dataset
The training dataset.
grad : list or numpy 1-D array
The value of the first order derivative (gradient) of the loss
with respect to the elements of preds for each sample point.
hess : list or numpy 1-D array
The value of the second order derivative (Hessian) of the loss
with respect to the elements of preds for each sample point.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is score[j * num_data + i]
and you should group grad and hess in this way as well.
feval : callable, list of callable, or None, optional (default=None)
Customized evaluation function.
Each evaluation function should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
preds : list or numpy 1-D array
The predicted values.
If ``fobj`` is specified, predicted values are returned before any transformation,
e.g. they are raw margin instead of probability of positive class for binary task in this case.
train_data : Dataset
The training dataset.
eval_name : str
The name of evaluation function (without whitespaces).
eval_result : float
The eval result.
is_higher_better : bool
Is eval result higher better, e.g. AUC is ``is_higher_better``.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
To ignore the default metric corresponding to the used objective,
set the ``metric`` parameter to the string ``"None"`` in ``params``.
init_model : str, pathlib.Path, Booster or None, optional (default=None)
Filename of LightGBM model or Booster instance used for continue training.
feature_name : list of str, or 'auto', optional (default="auto")
Feature names.
If 'auto' and data is pandas DataFrame, data columns names are used.
categorical_feature : list of str or int, or 'auto', optional (default="auto")
Categorical features.
If list of int, interpreted as indices.
If list of str, interpreted as feature names (need to specify ``feature_name`` as well).
If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
All values in categorical features should be less than int32 max value (2147483647).
Large values could be memory consuming. Consider using consecutive integers starting from zero.
All negative values in categorical features will be treated as missing values.
The output cannot be monotonically constrained with respect to a categorical feature.
early_stopping_rounds : int or None, optional (default=None)
Activates early stopping. The model will train until the validation score stops improving.
Validation score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue training.
Requires at least one validation data and one metric.
If there's more than one, will check all of them. But the training data is ignored anyway.
To check only the first metric, set the ``first_metric_only`` parameter to ``True`` in ``params``.
The index of iteration that has the best performance will be saved in the ``best_iteration`` field
if early stopping logic is enabled by setting ``early_stopping_rounds``.
evals_result : dict or None, optional (default=None)
Dictionary used to store all evaluation results of all the items in ``valid_sets``.
This should be initialized outside of your call to ``train()`` and should be empty.
Any initial contents of the dictionary will be deleted.
.. rubric:: Example
With a ``valid_sets`` = [valid_set, train_set],
``valid_names`` = ['eval', 'train']
and a ``params`` = {'metric': 'logloss'}
returns {'train': {'logloss': ['0.48253', '0.35953', ...]},
'eval': {'logloss': ['0.480385', '0.357756', ...]}}.
verbose_eval : bool or int, optional (default=True)
Requires at least one validation data.
If True, the eval metric on the valid set is printed at each boosting stage.
If int, the eval metric on the valid set is printed at every ``verbose_eval`` boosting stage.
The last boosting stage or the boosting stage found by using ``early_stopping_rounds`` is also printed.
.. rubric:: Example
With ``verbose_eval`` = 4 and at least one item in ``valid_sets``,
an evaluation metric is printed every 4 (instead of 1) boosting stages.
learning_rates : list, callable or None, optional (default=None)
List of learning rates for each boosting round
or a callable that calculates ``learning_rate``
in terms of current number of round (e.g. yields learning rate decay).
keep_training_booster : bool, optional (default=False)
Whether the returned Booster will be used to keep training.
If False, the returned value will be converted into _InnerPredictor before returning.
This means you won't be able to use ``eval``, ``eval_train`` or ``eval_valid`` methods of the returned Booster.
When your model is very large and cause the memory error,
you can try to set this param to ``True`` to avoid the model conversion performed during the internal call of ``model_to_string``.
You can still use _InnerPredictor as ``init_model`` for future continue training.
callbacks : list of callable, or None, optional (default=None)
List of callback functions that are applied at each iteration.
See Callbacks in Python API for more information.
Returns
-------
booster : Booster
The trained Booster model.
"""
# create predictor first
params = copy.deepcopy(params)
if fobj is not None:
for obj_alias in _ConfigAliases.get("objective"):
params.pop(obj_alias, None)
params['objective'] = 'none'
for alias in _ConfigAliases.get("num_iterations"):
if alias in params:
num_boost_round = params.pop(alias)
_log_warning(f"Found `{alias}` in params. Will use it instead of argument")
params["num_iterations"] = num_boost_round
# show deprecation warning only for early stop argument, setting early stop via global params should still be possible
if early_stopping_rounds is not None and early_stopping_rounds > 0:
_log_warning("'early_stopping_rounds' argument is deprecated and will be removed in a future release of LightGBM. "
"Pass 'early_stopping()' callback via 'callbacks' argument instead.")
for alias in _ConfigAliases.get("early_stopping_round"):
if alias in params:
early_stopping_rounds = params.pop(alias)
params["early_stopping_round"] = early_stopping_rounds
first_metric_only = params.get('first_metric_only', False)
if num_boost_round <= 0:
raise ValueError("num_boost_round should be greater than zero.")
predictor: Optional[_InnerPredictor] = None
if isinstance(init_model, (str, Path)):
predictor = _InnerPredictor(model_file=init_model, pred_parameter=params)
elif isinstance(init_model, Booster):
predictor = init_model._to_predictor(dict(init_model.params, **params))
init_iteration = predictor.num_total_iteration if predictor is not None else 0
# check dataset
if not isinstance(train_set, Dataset):
raise TypeError("Training only accepts Dataset object")
train_set._update_params(params) \
._set_predictor(predictor) \
.set_feature_name(feature_name) \
.set_categorical_feature(categorical_feature)
is_valid_contain_train = False
train_data_name = "training"
reduced_valid_sets = []
name_valid_sets = []
if valid_sets is not None:
if isinstance(valid_sets, Dataset):
valid_sets = [valid_sets]
if isinstance(valid_names, str):
valid_names = [valid_names]
for i, valid_data in enumerate(valid_sets):
# reduce cost for prediction training data
if valid_data is train_set:
is_valid_contain_train = True
if valid_names is not None:
train_data_name = valid_names[i]
continue
if not isinstance(valid_data, Dataset):
raise TypeError("Training only accepts Dataset object")
reduced_valid_sets.append(valid_data._update_params(params).set_reference(train_set))
if valid_names is not None and len(valid_names) > i:
name_valid_sets.append(valid_names[i])
else:
name_valid_sets.append(f'valid_{i}')
# process callbacks
if callbacks is None:
callbacks = set()
else:
for i, cb in enumerate(callbacks):
cb.__dict__.setdefault('order', i - len(callbacks))
callbacks = set(callbacks)
# Most of legacy advanced options becomes callbacks
if verbose_eval != "warn":
_log_warning("'verbose_eval' argument is deprecated and will be removed in a future release of LightGBM. "
"Pass 'log_evaluation()' callback via 'callbacks' argument instead.")
else:
if callbacks: # assume user has already specified log_evaluation callback
verbose_eval = False
else:
verbose_eval = True
if verbose_eval is True:
callbacks.add(callback.log_evaluation())
elif isinstance(verbose_eval, int):
callbacks.add(callback.log_evaluation(verbose_eval))
if early_stopping_rounds is not None and early_stopping_rounds > 0:
callbacks.add(callback.early_stopping(early_stopping_rounds, first_metric_only, verbose=bool(verbose_eval)))
if learning_rates is not None:
_log_warning("'learning_rates' argument is deprecated and will be removed in a future release of LightGBM. "
"Pass 'reset_parameter()' callback via 'callbacks' argument instead.")
callbacks.add(callback.reset_parameter(learning_rate=learning_rates))
if evals_result is not None:
_log_warning("'evals_result' argument is deprecated and will be removed in a future release of LightGBM. "
"Pass 'record_evaluation()' callback via 'callbacks' argument instead.")
callbacks.add(callback.record_evaluation(evals_result))
callbacks_before_iter = {cb for cb in callbacks if getattr(cb, 'before_iteration', False)}
callbacks_after_iter = callbacks - callbacks_before_iter
callbacks_before_iter = sorted(callbacks_before_iter, key=attrgetter('order'))
callbacks_after_iter = sorted(callbacks_after_iter, key=attrgetter('order'))
# construct booster
try:
booster = Booster(params=params, train_set=train_set)
if is_valid_contain_train:
booster.set_train_data_name(train_data_name)
for valid_set, name_valid_set in zip(reduced_valid_sets, name_valid_sets):
booster.add_valid(valid_set, name_valid_set)
finally:
train_set._reverse_update_params()
for valid_set in reduced_valid_sets:
valid_set._reverse_update_params()
booster.best_iteration = 0
# start training
for i in range(init_iteration, init_iteration + num_boost_round):
for cb in callbacks_before_iter:
cb(callback.CallbackEnv(model=booster,
params=params,
iteration=i,
begin_iteration=init_iteration,
end_iteration=init_iteration + num_boost_round,
evaluation_result_list=None))
booster.update(fobj=fobj)
evaluation_result_list = []
# check evaluation result.
if valid_sets is not None:
if is_valid_contain_train:
evaluation_result_list.extend(booster.eval_train(feval))
evaluation_result_list.extend(booster.eval_valid(feval))
try:
for cb in callbacks_after_iter:
cb(callback.CallbackEnv(model=booster,
params=params,
iteration=i,
begin_iteration=init_iteration,
end_iteration=init_iteration + num_boost_round,
evaluation_result_list=evaluation_result_list))
except callback.EarlyStopException as earlyStopException:
booster.best_iteration = earlyStopException.best_iteration + 1
evaluation_result_list = earlyStopException.best_score
break
booster.best_score = collections.defaultdict(collections.OrderedDict)
for dataset_name, eval_name, score, _ in evaluation_result_list:
booster.best_score[dataset_name][eval_name] = score
if not keep_training_booster:
booster.model_from_string(booster.model_to_string(), verbose='_silent_false').free_dataset()
return booster
[docs]class CVBooster:
"""CVBooster in LightGBM.
Auxiliary data structure to hold and redirect all boosters of ``cv`` function.
This class has the same methods as Booster class.
All method calls are actually performed for underlying Boosters and then all returned results are returned in a list.
Attributes
----------
boosters : list of Booster
The list of underlying fitted models.
best_iteration : int
The best iteration of fitted model.
"""
[docs] def __init__(self):
"""Initialize the CVBooster.
Generally, no need to instantiate manually.
"""
self.boosters = []
self.best_iteration = -1
def _append(self, booster):
"""Add a booster to CVBooster."""
self.boosters.append(booster)
def __getattr__(self, name):
"""Redirect methods call of CVBooster."""
def handler_function(*args, **kwargs):
"""Call methods with each booster, and concatenate their results."""
ret = []
for booster in self.boosters:
ret.append(getattr(booster, name)(*args, **kwargs))
return ret
return handler_function
def _make_n_folds(full_data, folds, nfold, params, seed, fpreproc=None, stratified=True,
shuffle=True, eval_train_metric=False):
"""Make a n-fold list of Booster from random indices."""
full_data = full_data.construct()
num_data = full_data.num_data()
if folds is not None:
if not hasattr(folds, '__iter__') and not hasattr(folds, 'split'):
raise AttributeError("folds should be a generator or iterator of (train_idx, test_idx) tuples "
"or scikit-learn splitter object with split method")
if hasattr(folds, 'split'):
group_info = full_data.get_group()
if group_info is not None:
group_info = np.array(group_info, dtype=np.int32, copy=False)
flatted_group = np.repeat(range(len(group_info)), repeats=group_info)
else:
flatted_group = np.zeros(num_data, dtype=np.int32)
folds = folds.split(X=np.empty(num_data), y=full_data.get_label(), groups=flatted_group)
else:
if any(params.get(obj_alias, "") in {"lambdarank", "rank_xendcg", "xendcg",
"xe_ndcg", "xe_ndcg_mart", "xendcg_mart"}
for obj_alias in _ConfigAliases.get("objective")):
if not SKLEARN_INSTALLED:
raise LightGBMError('scikit-learn is required for ranking cv')
# ranking task, split according to groups
group_info = np.array(full_data.get_group(), dtype=np.int32, copy=False)
flatted_group = np.repeat(range(len(group_info)), repeats=group_info)
group_kfold = _LGBMGroupKFold(n_splits=nfold)
folds = group_kfold.split(X=np.empty(num_data), groups=flatted_group)
elif stratified:
if not SKLEARN_INSTALLED:
raise LightGBMError('scikit-learn is required for stratified cv')
skf = _LGBMStratifiedKFold(n_splits=nfold, shuffle=shuffle, random_state=seed)
folds = skf.split(X=np.empty(num_data), y=full_data.get_label())
else:
if shuffle:
randidx = np.random.RandomState(seed).permutation(num_data)
else:
randidx = np.arange(num_data)
kstep = int(num_data / nfold)
test_id = [randidx[i: i + kstep] for i in range(0, num_data, kstep)]
train_id = [np.concatenate([test_id[i] for i in range(nfold) if k != i]) for k in range(nfold)]
folds = zip(train_id, test_id)
ret = CVBooster()
for train_idx, test_idx in folds:
train_set = full_data.subset(sorted(train_idx))
valid_set = full_data.subset(sorted(test_idx))
# run preprocessing on the data set if needed
if fpreproc is not None:
train_set, valid_set, tparam = fpreproc(train_set, valid_set, params.copy())
else:
tparam = params
cvbooster = Booster(tparam, train_set)
if eval_train_metric:
cvbooster.add_valid(train_set, 'train')
cvbooster.add_valid(valid_set, 'valid')
ret._append(cvbooster)
return ret
def _agg_cv_result(raw_results, eval_train_metric=False):
"""Aggregate cross-validation results."""
cvmap = collections.OrderedDict()
metric_type = {}
for one_result in raw_results:
for one_line in one_result:
if eval_train_metric:
key = f"{one_line[0]} {one_line[1]}"
else:
key = one_line[1]
metric_type[key] = one_line[3]
cvmap.setdefault(key, [])
cvmap[key].append(one_line[2])
return [('cv_agg', k, np.mean(v), metric_type[k], np.std(v)) for k, v in cvmap.items()]
[docs]def cv(params, train_set, num_boost_round=100,
folds=None, nfold=5, stratified=True, shuffle=True,
metrics=None, fobj=None, feval=None, init_model=None,
feature_name='auto', categorical_feature='auto',
early_stopping_rounds=None, fpreproc=None,
verbose_eval=None, show_stdv=True, seed=0,
callbacks=None, eval_train_metric=False,
return_cvbooster=False):
"""Perform the cross-validation with given parameters.
Parameters
----------
params : dict
Parameters for Booster.
train_set : Dataset
Data to be trained on.
num_boost_round : int, optional (default=100)
Number of boosting iterations.
folds : generator or iterator of (train_idx, test_idx) tuples, scikit-learn splitter object or None, optional (default=None)
If generator or iterator, it should yield the train and test indices for each fold.
If object, it should be one of the scikit-learn splitter classes
(https://scikit-learn.org/stable/modules/classes.html#splitter-classes)
and have ``split`` method.
This argument has highest priority over other data split arguments.
nfold : int, optional (default=5)
Number of folds in CV.
stratified : bool, optional (default=True)
Whether to perform stratified sampling.
shuffle : bool, optional (default=True)
Whether to shuffle before splitting data.
metrics : str, list of str, or None, optional (default=None)
Evaluation metrics to be monitored while CV.
If not None, the metric in ``params`` will be overridden.
fobj : callable or None, optional (default=None)
Customized objective function.
Should accept two parameters: preds, train_data,
and return (grad, hess).
preds : list or numpy 1-D array
The predicted values.
Predicted values are returned before any transformation,
e.g. they are raw margin instead of probability of positive class for binary task.
train_data : Dataset
The training dataset.
grad : list or numpy 1-D array
The value of the first order derivative (gradient) of the loss
with respect to the elements of preds for each sample point.
hess : list or numpy 1-D array
The value of the second order derivative (Hessian) of the loss
with respect to the elements of preds for each sample point.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is score[j * num_data + i]
and you should group grad and hess in this way as well.
feval : callable, list of callable, or None, optional (default=None)
Customized evaluation function.
Each evaluation function should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
preds : list or numpy 1-D array
The predicted values.
If ``fobj`` is specified, predicted values are returned before any transformation,
e.g. they are raw margin instead of probability of positive class for binary task in this case.
train_data : Dataset
The training dataset.
eval_name : str
The name of evaluation function (without whitespace).
eval_result : float
The eval result.
is_higher_better : bool
Is eval result higher better, e.g. AUC is ``is_higher_better``.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
To ignore the default metric corresponding to the used objective,
set ``metrics`` to the string ``"None"``.
init_model : str, pathlib.Path, Booster or None, optional (default=None)
Filename of LightGBM model or Booster instance used for continue training.
feature_name : list of str, or 'auto', optional (default="auto")
Feature names.
If 'auto' and data is pandas DataFrame, data columns names are used.
categorical_feature : list of str or int, or 'auto', optional (default="auto")
Categorical features.
If list of int, interpreted as indices.
If list of str, interpreted as feature names (need to specify ``feature_name`` as well).
If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
All values in categorical features should be less than int32 max value (2147483647).
Large values could be memory consuming. Consider using consecutive integers starting from zero.
All negative values in categorical features will be treated as missing values.
The output cannot be monotonically constrained with respect to a categorical feature.
early_stopping_rounds : int or None, optional (default=None)
Activates early stopping.
CV score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue.
Requires at least one metric. If there's more than one, will check all of them.
To check only the first metric, set the ``first_metric_only`` parameter to ``True`` in ``params``.
Last entry in evaluation history is the one from the best iteration.
fpreproc : callable or None, optional (default=None)
Preprocessing function that takes (dtrain, dtest, params)
and returns transformed versions of those.
verbose_eval : bool, int, or None, optional (default=None)
Whether to display the progress.
If True, progress will be displayed at every boosting stage.
If int, progress will be displayed at every given ``verbose_eval`` boosting stage.
show_stdv : bool, optional (default=True)
Whether to display the standard deviation in progress.
Results are not affected by this parameter, and always contain std.
seed : int, optional (default=0)
Seed used to generate the folds (passed to numpy.random.seed).
callbacks : list of callable, or None, optional (default=None)
List of callback functions that are applied at each iteration.
See Callbacks in Python API for more information.
eval_train_metric : bool, optional (default=False)
Whether to display the train metric in progress.
The score of the metric is calculated again after each training step, so there is some impact on performance.
return_cvbooster : bool, optional (default=False)
Whether to return Booster models trained on each fold through ``CVBooster``.
Returns
-------
eval_hist : dict
Evaluation history.
The dictionary has the following format:
{'metric1-mean': [values], 'metric1-stdv': [values],
'metric2-mean': [values], 'metric2-stdv': [values],
...}.
If ``return_cvbooster=True``, also returns trained boosters via ``cvbooster`` key.
"""
if not isinstance(train_set, Dataset):
raise TypeError("Training only accepts Dataset object")
params = copy.deepcopy(params)
if fobj is not None:
for obj_alias in _ConfigAliases.get("objective"):
params.pop(obj_alias, None)
params['objective'] = 'none'
for alias in _ConfigAliases.get("num_iterations"):
if alias in params:
_log_warning(f"Found `{alias}` in params. Will use it instead of argument")
num_boost_round = params.pop(alias)
params["num_iterations"] = num_boost_round
if early_stopping_rounds is not None and early_stopping_rounds > 0:
_log_warning("'early_stopping_rounds' argument is deprecated and will be removed in a future release of LightGBM. "
"Pass 'early_stopping()' callback via 'callbacks' argument instead.")
for alias in _ConfigAliases.get("early_stopping_round"):
if alias in params:
early_stopping_rounds = params.pop(alias)
params["early_stopping_round"] = early_stopping_rounds
first_metric_only = params.get('first_metric_only', False)
if num_boost_round <= 0:
raise ValueError("num_boost_round should be greater than zero.")
if isinstance(init_model, (str, Path)):
predictor = _InnerPredictor(model_file=init_model, pred_parameter=params)
elif isinstance(init_model, Booster):
predictor = init_model._to_predictor(dict(init_model.params, **params))
else:
predictor = None
if metrics is not None:
for metric_alias in _ConfigAliases.get("metric"):
params.pop(metric_alias, None)
params['metric'] = metrics
train_set._update_params(params) \
._set_predictor(predictor) \
.set_feature_name(feature_name) \
.set_categorical_feature(categorical_feature)
results = collections.defaultdict(list)
cvfolds = _make_n_folds(train_set, folds=folds, nfold=nfold,
params=params, seed=seed, fpreproc=fpreproc,
stratified=stratified, shuffle=shuffle,
eval_train_metric=eval_train_metric)
# setup callbacks
if callbacks is None:
callbacks = set()
else:
for i, cb in enumerate(callbacks):
cb.__dict__.setdefault('order', i - len(callbacks))
callbacks = set(callbacks)
if early_stopping_rounds is not None and early_stopping_rounds > 0:
callbacks.add(callback.early_stopping(early_stopping_rounds, first_metric_only, verbose=False))
if verbose_eval is not None:
_log_warning("'verbose_eval' argument is deprecated and will be removed in a future release of LightGBM. "
"Pass 'log_evaluation()' callback via 'callbacks' argument instead.")
if verbose_eval is True:
callbacks.add(callback.log_evaluation(show_stdv=show_stdv))
elif isinstance(verbose_eval, int):
callbacks.add(callback.log_evaluation(verbose_eval, show_stdv=show_stdv))
callbacks_before_iter = {cb for cb in callbacks if getattr(cb, 'before_iteration', False)}
callbacks_after_iter = callbacks - callbacks_before_iter
callbacks_before_iter = sorted(callbacks_before_iter, key=attrgetter('order'))
callbacks_after_iter = sorted(callbacks_after_iter, key=attrgetter('order'))
for i in range(num_boost_round):
for cb in callbacks_before_iter:
cb(callback.CallbackEnv(model=cvfolds,
params=params,
iteration=i,
begin_iteration=0,
end_iteration=num_boost_round,
evaluation_result_list=None))
cvfolds.update(fobj=fobj)
res = _agg_cv_result(cvfolds.eval_valid(feval), eval_train_metric)
for _, key, mean, _, std in res:
results[f'{key}-mean'].append(mean)
results[f'{key}-stdv'].append(std)
try:
for cb in callbacks_after_iter:
cb(callback.CallbackEnv(model=cvfolds,
params=params,
iteration=i,
begin_iteration=0,
end_iteration=num_boost_round,
evaluation_result_list=res))
except callback.EarlyStopException as earlyStopException:
cvfolds.best_iteration = earlyStopException.best_iteration + 1
for k in results:
results[k] = results[k][:cvfolds.best_iteration]
break
if return_cvbooster:
results['cvbooster'] = cvfolds
return dict(results)