Skip to content

Commit

Permalink
COSMIT rename n_iterations to n_iter in cross_validation
Browse files Browse the repository at this point in the history
  • Loading branch information
@amueller
amueller committed Oct 10, 2012
1 parent 1b367e5 commit 0a81827
Show file tree
Hide file tree
Showing 2 changed files with 44 additions and 36 deletions.
8 changes: 4 additions & 4 deletions doc/modules/cross_validation.rst
View file
Original file line number Diff line number Diff line change
Expand Up @@ -102,7 +102,7 @@ It is also possible to use othe cross validation strategies by passing a cross
validation iterator instead, for instance::

>>> n_samples = iris.data.shape[0]
>>> cv = cross_validation.ShuffleSplit(n_samples, n_iterations=3,
>>> cv = cross_validation.ShuffleSplit(n_samples, n_iter=3,
... test_size=0.3, random_state=0)

>>> cross_validation.cross_val_score(clf, iris.data, iris.target, cv=cv)
Expand Down Expand Up @@ -339,12 +339,12 @@ generator.

Here is a usage example::

>>> ss = cross_validation.ShuffleSplit(5, n_iterations=3, test_size=0.25,
>>> ss = cross_validation.ShuffleSplit(5, n_iter=3, test_size=0.25,
... random_state=0)
>>> len(ss)
3
>>> print ss # doctest: +ELLIPSIS
ShuffleSplit(5, n_iterations=3, test_size=0.25, indices=True, ...)
ShuffleSplit(5, n_iter=3, test_size=0.25, indices=True, ...)

>>> for train_index, test_index in ss:
... print train_index, test_index
Expand Down Expand Up @@ -390,7 +390,7 @@ smaller than the total dataset if it is very large.
>>> len(bs)
3
>>> print bs
Bootstrap(9, n_iterations=3, train_size=5, test_size=4, random_state=0)
Bootstrap(9, n_iter=3, train_size=5, test_size=4, random_state=0)

>>> for train_index, test_index in bs:
... print train_index, test_index
Expand Down
72 changes: 40 additions & 32 deletions sklearn/cross_validation.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -574,7 +574,7 @@ class Bootstrap(object):
"""Random sampling with replacement cross-validation iterator
Provides train/test indices to split data in train test sets
while resampling the input n_iterations times: each time a new
while resampling the input n_iter times: each time a new
random split of the data is performed and then samples are drawn
(with replacement) on each side of the split to build the training
and test sets.
Expand All @@ -592,7 +592,7 @@ class Bootstrap(object):
n : int
Total number of elements in the dataset.
n_iterations : int (default is 3)
n_iter : int (default is 3)
Number of bootstrapping iterations
train_size : int or float (default is 0.5)
Expand Down Expand Up @@ -623,7 +623,7 @@ class Bootstrap(object):
>>> len(bs)
3
>>> print(bs)
Bootstrap(9, n_iterations=3, train_size=5, test_size=4, random_state=0)
Bootstrap(9, n_iter=3, train_size=5, test_size=4, random_state=0)
>>> for train_index, test_index in bs:
... print("TRAIN: %s TEST: %s" % (train_index, test_index))
...
Expand All @@ -639,14 +639,14 @@ class Bootstrap(object):
# Static marker to be able to introspect the CV type
indices = True

def __init__(self, n, n_iterations=3, train_size=.5, test_size=None,
def __init__(self, n, n_iter=3, train_size=.5, test_size=None,
random_state=None, n_bootstraps=None):
self.n = n
if n_bootstraps is not None:
warnings.warn("n_bootstraps was renamed to n_iterations and will "
"be removed in 0.15.", DeprecationWarning)
n_iterations = n_bootstraps
self.n_iterations = n_iterations
warnings.warn("n_bootstraps was renamed to n_iter and will "
"be removed in 0.16.", DeprecationWarning)
n_iter = n_bootstraps
self.n_iter = n_iter
if (isinstance(train_size, (float, np.floating)) and train_size >= 0.0
and train_size <= 1.0):
self.train_size = ceil(train_size * n)
Expand Down Expand Up @@ -676,7 +676,7 @@ def __init__(self, n, n_iterations=3, train_size=.5, test_size=None,

def __iter__(self):
rng = check_random_state(self.random_state)
for i in range(self.n_iterations):
for i in range(self.n_iter):
# random partition
permutation = rng.permutation(self.n)
ind_train = permutation[:self.train_size]
Expand All @@ -691,18 +691,18 @@ def __iter__(self):
yield ind_train[train], ind_test[test]

def __repr__(self):
return ('%s(%d, n_iterations=%d, train_size=%d, test_size=%d, '
return ('%s(%d, n_iter=%d, train_size=%d, test_size=%d, '
'random_state=%d)' % (
self.__class__.__name__,
self.n,
self.n_iterations,
self.n_iter,
self.train_size,
self.test_size,
self.random_state,
))

def __len__(self):
return self.n_iterations
return self.n_iter


class ShuffleSplit(object):
Expand All @@ -719,7 +719,7 @@ class ShuffleSplit(object):
n : int
Total number of elements in the dataset.
n_iterations : int (default 10)
n_iter : int (default 10)
Number of re-shuffling & splitting iterations.
test_size : float (default 0.1) or int
Expand All @@ -744,21 +744,21 @@ class ShuffleSplit(object):
Examples
--------
>>> from sklearn import cross_validation
>>> rs = cross_validation.ShuffleSplit(4, n_iterations=3,
>>> rs = cross_validation.ShuffleSplit(4, n_iter=3,
... test_size=.25, random_state=0)
>>> len(rs)
3
>>> print(rs)
... # doctest: +ELLIPSIS
ShuffleSplit(4, n_iterations=3, test_size=0.25, indices=True, ...)
ShuffleSplit(4, n_iter=3, test_size=0.25, indices=True, ...)
>>> for train_index, test_index in rs:
... print("TRAIN: %s TEST: %s" % (train_index, test_index))
...
TRAIN: [3 1 0] TEST: [2]
TRAIN: [2 1 3] TEST: [0]
TRAIN: [0 2 1] TEST: [3]
>>> rs = cross_validation.ShuffleSplit(4, n_iterations=3,
>>> rs = cross_validation.ShuffleSplit(4, n_iter=3,
... train_size=0.5, test_size=.25, random_state=0)
>>> for train_index, test_index in rs:
... print("TRAIN: %s TEST: %s" % (train_index, test_index))
Expand All @@ -772,10 +772,14 @@ class ShuffleSplit(object):
Bootstrap: cross-validation using re-sampling with replacement.
"""

def __init__(self, n, n_iterations=10, test_size=0.1,
train_size=None, indices=True, random_state=None):
def __init__(self, n, n_iter=10, test_size=0.1, train_size=None,
indices=True, random_state=None, n_iterations=None):
self.n = n
self.n_iterations = n_iterations
self.n_iter = n_iter
if n_iterations is not None:
warnings.warn("n_iterations was renamed to n_iter for consistency "
" and will be removed in 0.16.")
self.n_iter = n_iterations
self.test_size = test_size
self.train_size = train_size
self.random_state = random_state
Expand All @@ -787,7 +791,7 @@ def __init__(self, n, n_iterations=10, test_size=0.1,

def __iter__(self):
rng = check_random_state(self.random_state)
for i in range(self.n_iterations):
for i in range(self.n_iter):
# random partition
permutation = rng.permutation(self.n)
ind_test = permutation[:self.n_test]
Expand All @@ -803,18 +807,18 @@ def __iter__(self):
yield train_mask, test_mask

def __repr__(self):
return ('%s(%d, n_iterations=%d, test_size=%s, indices=%s, '
return ('%s(%d, n_iter=%d, test_size=%s, indices=%s, '
'random_state=%s)' % (
self.__class__.__name__,
self.n,
self.n_iterations,
self.n_iter,
str(self.test_size),
self.indices,
self.random_state,
))

def __len__(self):
return self.n_iterations
return self.n_iter


def _validate_shuffle_split(n, test_size, train_size):
Expand Down Expand Up @@ -914,7 +918,7 @@ class StratifiedShuffleSplit(object):
y : array, [n_samples]
Labels of samples.
n_iterations : int (default 10)
n_iter : int (default 10)
Number of re-shuffling & splitting iterations.
test_size : float (default 0.1) or int
Expand Down Expand Up @@ -942,7 +946,7 @@ class StratifiedShuffleSplit(object):
>>> len(sss)
3
>>> print(sss) # doctest: +ELLIPSIS
StratifiedShuffleSplit(labels=[0 0 1 1], n_iterations=3, ...)
StratifiedShuffleSplit(labels=[0 0 1 1], n_iter=3, ...)
>>> for train_index, test_index in sss:
... print("TRAIN: %s TEST: %s" % (train_index, test_index))
... X_train, X_test = X[train_index], X[test_index]
Expand All @@ -952,12 +956,16 @@ class StratifiedShuffleSplit(object):
TRAIN: [0 2] TEST: [3 1]
"""

def __init__(self, y, n_iterations=10, test_size=0.1,
train_size=None, indices=True, random_state=None):
def __init__(self, y, n_iter=10, test_size=0.1, train_size=None,
indices=True, random_state=None, n_iterations=None):

self.y = np.array(y)
self.n = self.y.size
self.n_iterations = n_iterations
self.n_iter = n_iter
if n_iterations is not None:
warnings.warn("n_iterations was renamed to n_iter for consistency "
" and will be removed in 0.16.")
self.n_iter = n_iterations
self.test_size = test_size
self.train_size = train_size
self.random_state = random_state
Expand All @@ -973,7 +981,7 @@ def __iter__(self):
t_i = np.minimum(cls_count - n_i,
np.round(self.n_test * p_i).astype(int))

for n in range(self.n_iterations):
for n in range(self.n_iter):
train = []
test = []

Expand All @@ -998,18 +1006,18 @@ def __iter__(self):
yield train_m, test_m

def __repr__(self):
return ('%s(labels=%s, n_iterations=%d, test_size=%s, indices=%s, '
return ('%s(labels=%s, n_iter=%d, test_size=%s, indices=%s, '
'random_state=%s)' % (
self.__class__.__name__,
self.y,
self.n_iterations,
self.n_iter,
str(self.test_size),
self.indices,
self.random_state,
))

def __len__(self):
return self.n_iterations
return self.n_iter


##############################################################################
Expand Down

0 comments on commit 0a81827

Please sign in to comment.