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

Update Python docstring for ranking functions (#4121)

Browse Source 
* Update Python docstring for ranking functions

* Fix formatting
This commit is contained in:
Philip Hyunsu Cho 2019-02-10 12:22:02 -08:00 committed by GitHub
parent 3320a52192
commit 99a290489c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 41 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
python-package/xgboost/core.py
View File

@ -351,6 +351,14 @@ class DMatrix(object):
None, defaults to np.nan.
weight : list or numpy 1-D array , optional
Weight for each instance.
.. note:: For ranking task, weights are per-group.
In ranking task, one weight is assigned to each group (not each data
point). This is because we only care about the relative ordering of
data points within each group, so it doesn't make sense to assign
weights to individual data points.
silent : boolean, optional
Whether print messages during construction
feature_names : list, optional
@ -655,6 +663,13 @@ class DMatrix(object):
----------
weight : array like
Weight for each data point
.. note:: For ranking task, weights are per-group.
In ranking task, one weight is assigned to each group (not each data
point). This is because we only care about the relative ordering of
data points within each group, so it doesn't make sense to assign
weights to individual data points.
"""
self.set_float_info('weight', weight)
@ -666,6 +681,13 @@ class DMatrix(object):
----------
weight : array like
Weight for each data point in numpy 2D array
.. note:: For ranking task, weights are per-group.
In ranking task, one weight is assigned to each group (not each data
point). This is because we only care about the relative ordering of
data points within each group, so it doesn't make sense to assign
weights to individual data points.
"""
self.set_float_info_npy2d('weight', weight)

22
python-package/xgboost/sklearn.py
View File

@ -885,7 +885,7 @@ class XGBRanker(XGBModel):
Whether to print messages while running boosting.
objective : string
Specify the learning task and the corresponding learning objective.
Only "rank:pairwise" is supported currently.
The objective name must start with "rank:".
booster: string
Specify which booster to use: gbtree, gblinear or dart.
nthread : int
@ -999,13 +999,29 @@ class XGBRanker(XGBModel):
group : array_like
group size of training data
sample_weight : array_like
instance weights
group weights
.. note:: Weights are per-group for ranking tasks
In ranking task, one weight is assigned to each group (not each data
point). This is because we only care about the relative ordering of
data points within each group, so it doesn't make sense to assign
weights to individual data points.
eval_set : list, optional
A list of (X, y) tuple pairs to use as a validation set for
early-stopping
sample_weight_eval_set : list, optional
A list of the form [L_1, L_2, ..., L_n], where each L_i is a list of
instance weights on the i-th validation set.
group weights on the i-th validation set.
.. note:: Weights are per-group for ranking tasks
In ranking task, one weight is assigned to each group (not each data
point). This is because we only care about the relative ordering of
data points within each group, so it doesn't make sense to assign
weights to individual data points.
eval_group : list of arrays, optional
A list that contains the group size corresponds to each
(X, y) pair in eval_set