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

Fix error in Binomial to retain lazy logit initialization #46055

Closed
wants to merge 2 commits into from
Closed

Fix error in Binomial to retain lazy logit initialization #46055

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5688c47
Fix error in Binomial to retain lazy logit initialization
Oct 8, 2020
f792482
fix lint
Oct 8, 2020
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
5 changes: 3 additions & 2 deletions test/test_distributions.py
View file
Open in desktop
Expand Up @@ -4005,6 +4005,7 @@ def test_continuous_bernoulli_with_logits_overflow(self):
expected_gradient=tensor_type([0.]))


# TODO: make this a pytest parameterized test
class TestLazyLogitsInitialization(TestCase):
def setUp(self):
super(TestLazyLogitsInitialization, self).setUp()
Expand All @@ -4015,7 +4016,7 @@ def setUp(self):

def test_lazy_logits_initialization(self):
for Dist, params in self.examples:
param = params[0]
param = params[0].copy()
if 'probs' in param:
probs = param.pop('probs')
param['logits'] = probs_to_logits(probs)
Expand All @@ -4034,7 +4035,7 @@ def test_lazy_logits_initialization(self):

def test_lazy_probs_initialization(self):
for Dist, params in self.examples:
param = params[0]
param = params[0].copy()
if 'probs' in param:
dist = Dist(**param)
dist.sample()
Expand Down
10 changes: 2 additions & 8 deletions torch/distributions/binomial.py
View file
Open in desktop
@@ -1,4 +1,3 @@
from numbers import Number
import torch
from torch.distributions import constraints
from torch.distributions.distribution import Distribution
Expand Down Expand Up @@ -42,18 +41,13 @@ def __init__(self, total_count=1, probs=None, logits=None, validate_args=None):
raise ValueError("Either `probs` or `logits` must be specified, but not both.")
if probs is not None:
self.total_count, self.probs, = broadcast_all(total_count, probs)
self.total_count = self.total_count.type_as(self.logits)
is_scalar = isinstance(self.probs, Number)
self.total_count = self.total_count.type_as(self.probs)
else:
self.total_count, self.logits, = broadcast_all(total_count, logits)
self.total_count = self.total_count.type_as(self.logits)
is_scalar = isinstance(self.logits, Number)

self._param = self.probs if probs is not None else self.logits
if is_scalar:
batch_shape = torch.Size()
else:
batch_shape = self._param.size()
batch_shape = self._param.size()
super(Binomial, self).__init__(batch_shape, validate_args=validate_args)

def expand(self, batch_shape, _instance=None):
Expand Down