Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix evaluation result for XGBRanker. (#6594) (#6600)

Browse Source 
* Remove duplicated code, which fixes typo `evals_result` -> `evals_result_`.
This commit is contained in:
Jiaming Yuan 2021-01-13 04:42:43 +08:00 committed by GitHub
parent 8e321adac8
commit 6a29afb480
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 13 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
python-package/xgboost/sklearn.py
View File

@ -4,6 +4,7 @@
import copy import copy
import warnings import warnings
import json import json
from typing import Optional
import numpy as np import numpy as np
from .core import Booster, DMatrix, XGBoostError, _deprecate_positional_args from .core import Booster, DMatrix, XGBoostError, _deprecate_positional_args
from .training import train from .training import train
@ -494,6 +495,13 @@ class XGBModel(XGBModelBase):
# Delete the attribute after load # Delete the attribute after load
self.get_booster().set_attr(scikit_learn=None) self.get_booster().set_attr(scikit_learn=None)
def _set_evaluation_result(self, evals_result: Optional[dict]) -> None:
if evals_result:
for val in evals_result.items():
evals_result_key = list(val[1].keys())[0]
evals_result[val[0]][evals_result_key] = val[1][evals_result_key]
self.evals_result_ = evals_result
@_deprecate_positional_args @_deprecate_positional_args
def fit(self, X, y, *, sample_weight=None, base_margin=None, def fit(self, X, y, *, sample_weight=None, base_margin=None,
eval_set=None, eval_metric=None, early_stopping_rounds=None, eval_set=None, eval_metric=None, early_stopping_rounds=None,
@ -601,12 +609,7 @@ class XGBModel(XGBModelBase):
verbose_eval=verbose, xgb_model=xgb_model, verbose_eval=verbose, xgb_model=xgb_model,
callbacks=callbacks) callbacks=callbacks)
if evals_result: self._set_evaluation_result(evals_result)
for val in evals_result.items():
evals_result_key = list(val[1].keys())[0]
evals_result[val[0]][evals_result_key] = val[1][
evals_result_key]
self.evals_result_ = evals_result
if early_stopping_rounds is not None: if early_stopping_rounds is not None:
self.best_score = self._Booster.best_score self.best_score = self._Booster.best_score
@ -919,12 +922,7 @@ class XGBClassifier(XGBModel, XGBClassifierBase):
callbacks=callbacks) callbacks=callbacks)
self.objective = xgb_options["objective"] self.objective = xgb_options["objective"]
if evals_result: self._set_evaluation_result(evals_result)
for val in evals_result.items():
evals_result_key = list(val[1].keys())[0]
evals_result[val[0]][
evals_result_key] = val[1][evals_result_key]
self.evals_result_ = evals_result
if early_stopping_rounds is not None: if early_stopping_rounds is not None:
self.best_score = self._Booster.best_score self.best_score = self._Booster.best_score
@ -1328,12 +1326,7 @@ class XGBRanker(XGBModel):
self.objective = params["objective"] self.objective = params["objective"]
if evals_result: self._set_evaluation_result(evals_result)
for val in evals_result.items():
evals_result_key = list(val[1].keys())[0]
evals_result[val[0]][evals_result_key] = val[1][evals_result_key]
self.evals_result = evals_result
if early_stopping_rounds is not None: if early_stopping_rounds is not None:
self.best_score = self._Booster.best_score self.best_score = self._Booster.best_score
self.best_iteration = self._Booster.best_iteration self.best_iteration = self._Booster.best_iteration

2
tests/python/test_with_sklearn.py
View File

@ -122,6 +122,8 @@ def test_ranking():
model = xgb.sklearn.XGBRanker(**params) model = xgb.sklearn.XGBRanker(**params)
model.fit(x_train, y_train, group=train_group, model.fit(x_train, y_train, group=train_group,
eval_set=[(x_valid, y_valid)], eval_group=[valid_group]) eval_set=[(x_valid, y_valid)], eval_group=[valid_group])
assert model.evals_result()
pred = model.predict(x_test) pred = model.predict(x_test)
train_data = xgb.DMatrix(x_train, y_train) train_data = xgb.DMatrix(x_train, y_train)