Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fixed #1654: Fix mknfold using new StratifiedKFold API #1660

Merged
merged 1 commit into from Oct 12, 2016
Merged

Fixed #1654: Fix mknfold using new StratifiedKFold API #1660

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
4 changes: 2 additions & 2 deletions python-package/xgboost/compat.py
View file
Open in desktop
Expand Up @@ -46,12 +46,12 @@ class DataFrame(object):
try:
from sklearn.base import BaseEstimator
from sklearn.base import RegressorMixin, ClassifierMixin
from sklearn.preprocessing import LabelEncoder # noqa
from sklearn.preprocessing import LabelEncoder
try:
from sklearn.model_selection import KFold, StratifiedKFold
except ImportError:
from sklearn.cross_validation import KFold, StratifiedKFold

SKLEARN_INSTALLED = True

XGBModelBase = BaseEstimator
Expand Down
8 changes: 3 additions & 5 deletions python-package/xgboost/training.py
View file
Open in desktop
Expand Up @@ -232,14 +232,12 @@ def mknfold(dall, nfold, param, seed, evals=(), fpreproc=None, stratified=False,
randidx = np.random.permutation(dall.num_row())
kstep = int(len(randidx) / nfold)
idset = [randidx[(i * kstep): min(len(randidx), (i + 1) * kstep)] for i in range(nfold)]
elif folds is not None:
elif folds is not None and isinstance(folds, list):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this check blocks off KFold usage

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I commented on your PR. Could you address this in your PR too?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sure, thanks for the quick reply

idset = [x[1] for x in folds]
nfold = len(idset)
else:
idset = [x[1] for x in XGBStratifiedKFold(dall.get_label(),
n_folds=nfold,
shuffle=True,
random_state=seed)]
sfk = XGBStratifiedKFold(n_splits=nfold, shuffle=True, random_state=seed)
idset = [x[1] for x in sfk.split(X=dall.get_label(), y=dall.get_label())]

ret = []
for k in range(nfold):
Expand Down
6 changes: 3 additions & 3 deletions tests/python/test_with_sklearn.py
View file
Open in desktop
Expand Up @@ -251,7 +251,7 @@ def test_sklearn_plotting():
def test_sklearn_nfolds_cv():
tm._skip_if_no_sklearn()
from sklearn.datasets import load_digits
from sklearn.cross_validation import StratifiedKFold
from sklearn.model_selection import StratifiedKFold

digits = load_digits(3)
X = digits['data']
Expand All @@ -269,10 +269,10 @@ def test_sklearn_nfolds_cv():

seed = 2016
nfolds = 5
skf = StratifiedKFold(y, n_folds=nfolds, shuffle=True, random_state=seed)
skf = StratifiedKFold(n_splits=nfolds, shuffle=True, random_state=seed)

cv1 = xgb.cv(params, dm, num_boost_round=10, nfold=nfolds, seed=seed)
cv2 = xgb.cv(params, dm, num_boost_round=10, folds=skf, seed=seed)
cv2 = xgb.cv(params, dm, num_boost_round=10, nfold=nfolds, folds=skf, seed=seed)
cv3 = xgb.cv(params, dm, num_boost_round=10, nfold=nfolds, stratified=True, seed=seed)
assert cv1.shape[0] == cv2.shape[0] and cv2.shape[0] == cv3.shape[0]
assert cv2.iloc[-1, 0] == cv3.iloc[-1, 0]
Expand Down