-
Notifications
You must be signed in to change notification settings - Fork 74k
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.
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
TensorArray grad bug #13355
Comments
@alextp would you please take a look? |
@ebrevdo , can you take a look? |
This is definitely a bug, and it has to do with this line:
the weird "gradients" prefix is there so that the TensorArray can differentiate between different calls to Note that this will work: tf.gradients(y, x, [[[2.42]]]) because the grad_ys [[[2.42]]] is converted to a tensor inside the tf.gradients, and will have the appropriate name scope. One solution might be to wrap all grad_ys going into |
I have a solution; will test and push - if all is green, you will see it in 2-3 days. |
This is not fixed yet in 1.4.0. The test does not pass. |
Have you tried in the nightlies?
…On Fri, Nov 3, 2017 at 3:30 AM, Albert Zeyer ***@***.***> wrote:
This is not fixed yet in 1.4.0. The test does not pass.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#13355 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABtim6SW6yBKxW8Y-V7cS0vy6yw7CeBtks5syutCgaJpZM4Pm22x>
.
|
No, the official 1.4.0 release, installed via pip install. I thought that the commit should be in that release but not sure. Specifically, my version is |
Commits to master don't always make it into 1.4 releases, they have to be
cherrypicked during the RC and it's possible that this bugfix was not.
…On Fri, Nov 3, 2017 at 9:03 AM, Albert Zeyer ***@***.***> wrote:
No, the official 1.4.0 release, installed via pip install. I thought that
the commit should be in that release but not sure. Specifically, my version
is v1.4.0-rc1-11-g130a514 1.4.0.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#13355 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABtim5_uz7AWhBuBuqT_8fD9y3r8uglRks5syzlVgaJpZM4Pm22x>
.
|
System information
Describe the problem
tf.TensorArray
in some cases does not correctly passes the gradient. See the test case.Source code / logs
This fails:
I get the output:
Strangely, if you add something like
y = y[::1]
before taking the gradient, it passes.The text was updated successfully, but these errors were encountered: