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

Added an option to use the pure python implementation. #1137

Merged
merged 17 commits into from
Mar 7, 2020
Merged

Added an option to use the pure python implementation. #1137

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
e4f4fd7
Added an option to use the pure python implementation.
gabrieldemarmiesse Feb 23, 2020
91fb3a2
Forgot to add the new python file to the BUILD.
gabrieldemarmiesse Feb 23, 2020
ed123be
Some info.
gabrieldemarmiesse Feb 23, 2020
a17572e
Merge branch 'dodu' into enable_py_func_publicly
gabrieldemarmiesse Feb 23, 2020
44e2b73
Add options as dependency
seanpmorgan Feb 24, 2020
e717d0b
Lint
seanpmorgan Feb 24, 2020
73c149e
Merge branch 'master' into enable_py_func_publicly
gabrieldemarmiesse Feb 25, 2020
b66e1ac
Added warning and fallback.
gabrieldemarmiesse Feb 25, 2020
39ef6f8
Format.
gabrieldemarmiesse Feb 25, 2020
d8bc287
Format again.
gabrieldemarmiesse Feb 25, 2020
ab32156
Removed the option module from checks.
gabrieldemarmiesse Feb 25, 2020
fe9246f
Merge branch 'master' into enable_py_func_publicly
gabrieldemarmiesse Feb 26, 2020
60c7c9f
Merge branch 'master' into enable_py_func_publicly
gabrieldemarmiesse Feb 26, 2020
d6a3aac
Some rewording.
gabrieldemarmiesse Feb 26, 2020
3b906e2
Update README.md
gabrieldemarmiesse Feb 27, 2020
3183583
Update README.md
gabrieldemarmiesse Feb 27, 2020
415f1f3
Update README.md
gabrieldemarmiesse Feb 27, 2020
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
Prev Previous commit
Next Next commit
Format again.
  • Loading branch information
@gabrieldemarmiesse
gabrieldemarmiesse committed Feb 25, 2020
commit d8bc2871bed0e7ed4148ff28d4a4328a9f5dc3dc
2 changes: 0 additions & 2 deletions tensorflow_addons/activations/hardshrink.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,8 +13,6 @@
# limitations under the License.
# ==============================================================================

import warnings

import tensorflow as tf
from tensorflow_addons.utils.types import Number

Expand Down
14 changes: 7 additions & 7 deletions tensorflow_addons/options.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,18 +14,18 @@

FALLBACK_WARNING_TEMPLATE = """{}

The {} C++/CUDA custom op could not be loaded.
For this reason, Addons will fallback to an implementation written
The {} C++/CUDA custom op could not be loaded.
For this reason, Addons will fallback to an implementation written
in Python with public TensorFlow ops. There worst you might experience with
this is a moderate slowdown on GPU. There can be multiple
reason for this loading error, one of them may be an ABI incompatibility between
this is a moderate slowdown on GPU. There can be multiple
gabrieldemarmiesse marked this conversation as resolved.
Show resolved Hide resolved
reason for this loading error, one of them may be an ABI incompatibility between
the TensorFlow installed on your system and the TensorFlow used to compile
TensorFlow Addons' custom ops. The stacktrace generated when loading the
TensorFlow Addons' custom ops. The stacktrace generated when loading the
shared object file was displayed above.

If you want this warning to disappear, either make sure the TensorFlow installed
is compatible with this version of Addons, or tell TensorFlow Addons to
prefer using Python implementations and not custom C++/CUDA ones. You can do that
is compatible with this version of Addons, or tell TensorFlow Addons to
prefer using Python implementations and not custom C++/CUDA ones. You can do that
by changing the TF_ADDONS_PY_OPS flag
either with the environment variable:
```bash
Expand Down