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

add rrelu kernel #573

Merged
merged 14 commits into from
Oct 25, 2019
Merged

add rrelu kernel #573

Changes from 1 commit
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
7874b97
add rrelu kernel
fsx950223 Oct 6, 2019
3db6502
format code
fsx950223 Oct 7, 2019
bb26705
change test case
fsx950223 Oct 8, 2019
9f2eb56
simplify kernel implemention
fsx950223 Oct 9, 2019
e6e2815
fix test case
fsx950223 Oct 9, 2019
47aacae
refact test case
fsx950223 Oct 9, 2019
b1410c6
remove forward training test
fsx950223 Oct 11, 2019
736a3dd
format code
fsx950223 Oct 12, 2019
c75136b
modify test case
fsx950223 Oct 16, 2019
e708056
add forget files
fsx950223 Oct 20, 2019
d0c117d
update kernel random implementation
fsx950223 Oct 21, 2019
c24995c
fix rrelu test bug
fsx950223 Oct 25, 2019
41f1633
add TODO
fsx950223 Oct 25, 2019
acf4374
fix comment
fsx950223 Oct 25, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 9 additions & 1 deletion tensorflow_addons/activations/rrelu_test.py
100644 → 100755
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,15 @@ def test_theoretical_gradients(self, dtype):
for training in [True, False]:
with self.subTest(training=training):
theoretical, numerical = tf.test.compute_gradient(
lambda x: rrelu(x, lower, upper, training=training), [x])
lambda x: rrelu(
x, lower, upper, training=training, seed=111111), [x])
if training is True and tf.test.is_gpu_available() is False:
fsx950223 marked this conversation as resolved.
Show resolved Hide resolved
numerical = [[[0.134971, 0., 0., 0., 0., 0.],
[0., 0.15648358, 0., 0., 0., 0.],
[0., 0., 0.18776372, 0., 0., 0.],
[0., 0., 0., 1., 0., 0.],
[0., 0., 0., 0., 1., 0.],
[0., 0., 0., 0., 0., 1.]]]
self.assertAllCloseAccordingToType(
theoretical, numerical, rtol=5e-4, atol=5e-4)

Expand Down