profile
viewpoint
Gabriel de Marmiesse gabrieldemarmiesse Valeo.ai Paris, France Research engineer, deep learning

gabrieldemarmiesse/python-autopxd2 13

A friendly fork of autopxd

gabrieldemarmiesse/Allstate-claims-severity-competition 3

A pyspark Ipython notebook on the Kaggle competition.

gabrieldemarmiesse/getting_started_open_source 3

You want to contribute to an open-source project? You don't know how to do it? Here is how to.

gabrieldemarmiesse/hualos 2

Keras Total Visualization project

gabrieldemarmiesse/AdaptSegNet 1

Learning to Adapt Structured Output Space for Semantic Segmentation, CVPR 2018 (spotlight)

gabrieldemarmiesse/cython 1

The most widely used Python to C compiler

gabrieldemarmiesse/KerasToCPP 1

This is a library that convert Keras models into binaries that a C++ program can read.

pull request commenttensorflow/addons

remove pip cache

Can you try docker builder prune --filter type=exec.cachemount?

autoih

comment created time in 18 minutes

pull request commenttensorflow/addons

remove pip cache

My bad, I thought it was something else. I don't see why this problem is happening, I can't reproduce the bug locally. It's not happening in the CI either. We can't merge this fix until we know what's happening and if this is an isolated problem or not. We need more info.

autoih

comment created time in 23 minutes

pull request commentkeras-team/keras-autodoc

Fix support for variable-size indentation (e.g. 2-space indentation)

Cool thanks a lot! Maybe we could have a minimal test?

fchollet

comment created time in 31 minutes

pull request commenttensorflow/addons

remove pip cache

The idea is that this syntax is only working when using docker buildkit. Docker buildkit is available as experimental feature for docker 19.03+. To activate it, you just need to set the environement variable DOCKER_BUILDKIT=1. We already do that in the bash scripts we use in the CI, but it seems you tried to run the docker build directly. Can you run DOCKER_BUILDKIT=1 docker build ...? It should work.

autoih

comment created time in 41 minutes

pull request commenttensorflow/addons

remove pip cache

Sure thing, that's helpful, can you also do docker info? This command is important.

autoih

comment created time in an hour

pull request commenttensorflow/addons

remove pip cache

Can you do docker info and paste the result here?

autoih

comment created time in an hour

pull request commenttensorflow/addons

remove pip cache

@autoih is it an error that you get locally or in the CI?

autoih

comment created time in an hour

pull request commenttensorflow/addons

Remove the pip cache in the manylinux image. Fix the master branch build.

It seems that it won't work, even if we don't install the pytest dependencies (scikit-image and such).

gabrieldemarmiesse

comment created time in 3 hours

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha d7c2a7843f47414e1c3cbde9143aaf0cbcbf0492

Take 2

view details

push time in 3 hours

pull request commenttensorflow/addons

Moved test out of run_in_graph_and_eager_mode in triplet_test

If you scroll in the logs, you'll see the stacktrace. It looks like a flaky test. We should have increased the absolute and relative tolerance to 1e-6. Also be aware that the build of the master branch is currently broken because we use too much disk space. So you may experience failed builds because of that until we fix it.

autoih

comment created time in 4 hours

create barnchgabrieldemarmiesse/addons

branch : reduce_size_build

created branch time in 4 hours

pull request commenttensorflow/addons

Users don't need to declare the tests now in the image submodule.

Pro tip: To avoid the bot spamming everyone, do the modifications in the second commit :p

gabrieldemarmiesse

comment created time in 4 hours

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha b4cf4b6436e0198b8d5231b671cb2de7f1c6edbf

Other stuff to avoid the bot calling everyone.

view details

push time in 4 hours

create barnchgabrieldemarmiesse/addons

branch : cut_build_image

created branch time in 4 hours

Pull request review commenttensorflow/addons

Removed run_in_graph_and_eager_mode in npairs_test

  import sys import platform-import unittest +import numpy as np import pytest import tensorflow as tf from tensorflow_addons.losses import npairs-from tensorflow_addons.utils import test_utils  IS_WINDOWS = platform.system() == "Windows"  -@unittest.skipIf(+@pytest.mark.skipif(

rather than using pytest.mark.skipif all the time, you can use pytestmark = pytest.mark.skipif(...) at the top of the file. See http://doc.pytest.org/en/latest/example/markers.html#marking-whole-classes-or-modules

autoih

comment created time in 9 hours

push eventtensorflow/addons

Jhuo IH

commit sha 8cf744845030a83f0179a3dc6a8dee210eaeda1f

Removed run_in_graph_and_eager_mode in contrastive_test (#1519)

view details

push time in 10 hours

PR merged tensorflow/addons

Reviewers
Removed run_in_graph_and_eager_mode in contrastive_test awaiting testing (then merge) cla: yes losses

cc @gabrieldemarmiesse

+53 -57

0 comment

1 changed file

autoih

pr closed time in 10 hours

push eventtensorflow/addons

Jeremy Oldfather

commit sha eca9f12b054aa298fee2007790d165b5c61e6445

Fix unhashable decay_var_list (#1510) * Fix unhashable decay_var_list * Ran black ./ before commit * Removed backward compatibility with TF 1.x * Added test cases to check decay_var_list Added test cases for both SGDW and AdamW * Ran black ./

view details

push time in a day

PR merged tensorflow/addons

Fix unhashable decay_var_list cla: yes optimizers

Fixes an error where the list of decay variables is unhashable because they are compared without var.ref() or var.experimental_ref(). This patch is forward compatible with TF 2.2.0.

+56 -7

4 comments

2 changed files

theoldfather

pr closed time in a day

push eventtensorflow/addons

Jhuo IH

commit sha 205bdadeb373da956e7dbc568d87c2867a53b2c7

Removed run_in_graph_and_eager_mode in normalization_test (#1518)

view details

push time in a day

PR merged tensorflow/addons

Removed run_in_graph_and_eager_mode in normalization_test cla: yes layers

cc @gabrieldemarmiesse. This may complete the normalization test.

+65 -72

1 comment

1 changed file

autoih

pr closed time in a day

push eventtensorflow/addons

Jhuo IH

commit sha fc64cbb929e5ef2078e83612da0220479c05f0e9

Move test out of run_all_in_graph_and_eager_mode in normalizations_test (#1517)

view details

push time in a day

PR merged tensorflow/addons

Move test out of run_all_in_graph_and_eager_mode in normalizations_test awaiting testing (then merge) cla: yes layers

cc @gabrieldemarmiesse

+50 -48

1 comment

1 changed file

autoih

pr closed time in a day

push eventtensorflow/addons

Jhuo IH

commit sha 0f477c61880de2465bfb3e5375b5a69f1509c20c

Moved tests out of run_in_graph_and_eager_mode in wrapers_test (#1516)

view details

push time in a day

PR merged tensorflow/addons

Reviewers
Moved tests out of run_in_graph_and_eager_mode in wrapers_test awaiting testing (then merge) cla: yes layers

cc @gabrieldemarmiesse

+48 -41

0 comment

1 changed file

autoih

pr closed time in a day

create barnchgabrieldemarmiesse/addons

branch : fix_wrapper

created branch time in a day

push eventtensorflow/addons

Jhuo IH

commit sha 8749ffd80cea28fb77bb5dfc8cf46cd86e05b4c2

Moved test out run_in_graph_and_eager_mode in wrapper (#1513)

view details

push time in a day

PR merged tensorflow/addons

Reviewers
Moved test out run_in_graph_and_eager_mode in wrapper awaiting testing (then merge) cla: yes layers

cc @gabrieldemarmiesse

+55 -40

0 comment

1 changed file

autoih

pr closed time in a day

push eventtensorflow/addons

Jhuo IH

commit sha 30282a58c831628bad942d7c5b65fc8f324fdcae

Moved tests out of run_in_graph_and_eager_mode in wrappers (#1514)

view details

push time in a day

PR merged tensorflow/addons

Reviewers
Moved tests out of run_in_graph_and_eager_mode in wrappers awaiting testing (then merge) cla: yes layers

cc @gabrieldemarmiesse

+27 -26

0 comment

1 changed file

autoih

pr closed time in a day

pull request commenttensorflow/addons

Fix unhashable decay_var_list

@theoldfather thank you! I'll take a look.

theoldfather

comment created time in a day

pull request commenttensorflow/addons

Add a warning when trying to load a custom op with a uncompatible version of TF.

I believe we can add more info in the readme. Let's not bload the warning too much, it's already very descriptive.

gabrieldemarmiesse

comment created time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha cce70a6240ae887069dab1c71025e4ea2c09494f

Small rewording.

view details

push time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha 7668b0fc44b993c4c9816f80744a9954a504f018

Update message.

view details

push time in a day

issue commenttensorflow/addons

TimeStopping callback is untested.

We should just have a fit() call and then we check that the callback is doing what we expect it to do. For the TimeStopping callback, we can check the running time of the fit call for example.

gabrieldemarmiesse

comment created time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha 1074573ee6496748bf47bc05dd7fedc25dfbdc47

Forgot to set variable.

view details

push time in a day

create barnchgabrieldemarmiesse/addons

branch : easier_python_management

created branch time in a day

issue commenttensorflow/addons

Make our CI build wheels against TF 2.1 and 2.2

What is currently missing is the docker image to build the manylinux wheels for 2.2.0. It's likely that this docker image will be available after the stable release.

gabrieldemarmiesse

comment created time in a day

pull request commenttensorflow/addons

Moved test out of run_in_eager_and_graph_mode in resampler_ops.

The cla is not preventing us from merging. So no worries on this part.

gabrieldemarmiesse

comment created time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha 8bd90eb840e12155e42914477d88e32b23d6e6ba

Removed unused import.

view details

push time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha fa68c926363cbc37417c814fb983fe750b2ee65c

Removed boilerplate.

view details

push time in a day

pull request commenttensorflow/addons

Added a minimal check to ensure we don't use tf private API.

In the end I grouped all tests together and run them with pytest. I believe it's cleaner and it's going to be easier to add new checks from #1279

gabrieldemarmiesse

comment created time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha b023f093c54610f5406cea6d1ecd176e6567978f

Change the path in contributing.

view details

push time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha d575ea07ca18b98fbbd0b28e796e2ec526adb030

Remove no-deps.

view details

push time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha 20122d143673df278dacbb7d02b0703157222e91

Group stuff together.

view details

push time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha ce0d155891168dd47083219a6b84ab74c3410477

Run pytest on the directory.

view details

push time in a day

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha ad8ee92c37c8f5c4f7889826598274b8053b809c

Group tests together.

view details

push time in a day

push eventtensorflow/addons

Jhuo IH

commit sha 874c98882c0850e6a0a36b029a9130ab617d7c46

Move test out of run_all_in_graph_and_eager_mode in normalizations_test (#1511)

view details

push time in a day

PR merged tensorflow/addons

Move test out of run_all_in_graph_and_eager_mode in normalizations_test cla: yes layers

cc @gabrieldemarmiesse

+23 -20

1 comment

1 changed file

autoih

pr closed time in a day

pull request commenttensorflow/addons

Fix unhashable decay_var_list

@theoldfather could you provide a test in your pull request to ensure that we don't re-introduce the bug in the future?

Ideally this test should fail without your patch.

theoldfather

comment created time in a day

push eventtensorflow/addons

Jhuo IH

commit sha 2e4dc869fb88172deff1018890c4fa6af9eb16a2

Moved tests out of run_in_graph_and_eager_mode in normalizations (#1508)

view details

push time in 2 days

PR merged tensorflow/addons

Moved tests out of run_in_graph_and_eager_mode in normalizations awaiting testing (then merge) cla: yes layers

cc @gabrieldemarmiesse

+28 -26

1 comment

1 changed file

autoih

pr closed time in 2 days

issue closedtensorflow/addons

tensorflow2.0 can't use this module

System information

  • OS Platform and Distribution (e.g., Linux Ubuntu 16.04): window10
  • TensorFlow version and how it was installed (source or binary): 2.0.0
  • TensorFlow-Addons version and how it was installed (source or binary): get it by github
  • Python version: 3.6
  • Is GPU used? (yes/no): no

Describe the bug when i use addons in my program ,it tell me that i only use this module in tensorflow2.1.0, Do you know how can get this module be used in tensorflow2.0.0

closed time in 2 days

ididhmc

issue commenttensorflow/addons

tensorflow2.0 can't use this module

Closing this issue since we got no answer from the author. Tracking the docs part in #1317

ididhmc

comment created time in 2 days

PR opened tensorflow/addons

Added a minimal check to ensure we don't use tf private API.

It won't catch all the usages of tensorflow's private API, nor it's 100% accurate, but it should help by removing some work off the shoulders of the reviewers.

+48 -1

0 comment

2 changed files

pr created time in 2 days

create barnchgabrieldemarmiesse/addons

branch : no_private_API

created branch time in 2 days

PR closed tensorflow/addons

[do not merge] Try enable all cla: yes github
+14 -8

2 comments

1 changed file

gabrieldemarmiesse

pr closed time in 2 days

pull request commenttensorflow/addons

[do not merge] Try enable all

I don't know why this is not working :( We can do more tests later on.

gabrieldemarmiesse

comment created time in 2 days

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha e2447b23b3231134b684134e1f466803997debf2

Remove ubuntu exception.

view details

push time in 2 days

push eventtensorflow/addons

Jhuo IH

commit sha 201ad30470f36cc28b2e94117d1d557c3cada590

Moved tests out of run_in_graph_and_eager_mode in sparsemax_test (#1505)

view details

push time in 2 days

PR merged tensorflow/addons

Moved tests out of run_in_graph_and_eager_mode in sparsemax_test awaiting testing (then merge) cla: yes layers

cc @gabrieldemarmiesse

+12 -15

1 comment

1 changed file

autoih

pr closed time in 2 days

pull request commenttensorflow/addons

Add python 3.8 to the build matrix

It seems to work. You can take a look at https://github.com/tensorflow/addons/pull/1506 which is the same PR but the CI is building everything.

gabrieldemarmiesse

comment created time in 2 days

PR opened tensorflow/addons

Try enable all
+18 -8

0 comment

1 changed file

pr created time in 2 days

create barnchgabrieldemarmiesse/addons

branch : try_enable_all

created branch time in 2 days

PR closed tensorflow/addons

Trying the 38 setup cla: yes github
+5 -5

0 comment

1 changed file

gabrieldemarmiesse

pr closed time in 2 days

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha 0421401a022cadec1fa3c08f38590a39f47912f5

Stuff

view details

push time in 2 days

push eventtensorflow/addons

Jhuo IH

commit sha 45f96d96ab15bf3939d9867ea0a1f2a7d8c395ad

Moved tests out of run_in_graph_and_eager_mode in sparse_image_warp_test (#1477) * Moved tests out of run_in_graph_and_eager_mode in sparse_image_warp_test * snake case * minor fix * test_zero_shift * messages

view details

push time in 2 days

PR merged tensorflow/addons

Moved tests out of run_in_graph_and_eager_mode in sparse_image_warp_test cla: yes image

cc @gabrieldemarmiesse

+21 -24

1 comment

1 changed file

autoih

pr closed time in 2 days

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha 0dc6c7b0f0b5aa17db84cbe58669c1a4bcf869f8

Multiline for all.

view details

push time in 2 days

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha a3b7ac1c537e3182a7c3df0e567fd978078324b2

New syntax.

view details

push time in 2 days

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha 66edbd27e6b3ac217488b4ddf2b9114b453eb933

Try multiline.

view details

push time in 2 days

PR opened tensorflow/addons

Trying the 38 setup
+5 -5

0 comment

1 changed file

pr created time in 2 days

create barnchgabrieldemarmiesse/addons

branch : trying_the_38_setup

created branch time in 2 days

PR opened tensorflow/addons

Add python 3.8 to the build matrix
+5 -5

0 comment

1 changed file

pr created time in 2 days

create barnchgabrieldemarmiesse/addons

branch : add_py38

created branch time in 2 days

issue openedtensorflow/addons

AverageModelCheckpoint is untested

We have no way of knowing if it's working or not. Some tests would ensure no bugs are introduced.

created time in 2 days

issue openedtensorflow/addons

TimeStopping callback is untested.

We have no way of knowing if it works or not. Some minimal tests would be greatly appreciated.

created time in 2 days

issue openedtensorflow/addons

TQDMProgressBar needs some tests.

It's currently untested. We have no way of knowing if this class works or not.

See also #1495

created time in 2 days

pull request commenttensorflow/addons

Moved tests out of run_in_graph_and_eager_mode in normalizatoin_test

Thanks again, that's a great help, if I had to do it alone, it would take forever!

autoih

comment created time in 2 days

push eventtensorflow/addons

Jhuo IH

commit sha 51ab148e97002f9003de65e4dac1b45df8113504

Moved tests out of run_in_graph_and_eager_mode in normalizatoin_test (#1498)

view details

push time in 2 days

PR merged tensorflow/addons

Moved tests out of run_in_graph_and_eager_mode in normalizatoin_test awaiting testing (then merge) cla: yes layers

cc @gabrieldemarmiesse

+29 -25

1 comment

1 changed file

autoih

pr closed time in 2 days

PR opened tensorflow/addons

Reviewers
Remove the need to declare the test in the build file.

Given that all the tensorflow_addons/activations tests runs under 3 seconds, it's maybe not needed for bazel users to select exactly which one they want to run.

+10 -101

0 comment

3 changed files

pr created time in 2 days

create barnchgabrieldemarmiesse/addons

branch : test_all_in_dir

created branch time in 2 days

Pull request review commenttensorflow/addons

Add tf 2.2.0rc1 in the build matrix.

 ARG PY_VERSION RUN python$PY_VERSION -m pip install --upgrade pip setuptools auditwheel==2.0.0  COPY tools/install_deps/ /install_deps+ARG TF_VERSION

See https://docs.docker.com/engine/reference/builder/#understand-how-arg-and-from-interact

gabrieldemarmiesse

comment created time in 2 days

Pull request review commenttensorflow/addons

Add tf 2.2.0rc1 in the build matrix.

 ARG PY_VERSION RUN python$PY_VERSION -m pip install --upgrade pip setuptools auditwheel==2.0.0  COPY tools/install_deps/ /install_deps+ARG TF_VERSION

When using an ARG before a from, the ARG is not valid afterwards and needs to be declared again from what I remember. Let me find a piece of docs about that.

gabrieldemarmiesse

comment created time in 2 days

Pull request review commenttensorflow/addons

Add tf 2.2.0rc1 in the build matrix.

 jobs:     strategy:       matrix:         os: ['macos-latest', 'windows-latest', 'ubuntu-18.04']-        python-version: ['3.5', '3.6', '3.7']+        py-version: ['3.5', '3.6', '3.7']+        tf-version: ['2.1.0', '2.2.0rc1']       fail-fast: false     steps:       - if: github.event_name == 'push'         run: echo "::set-env name=NIGHTLY_FLAG::--nightly"       - uses: actions/checkout@v2       - uses: actions/setup-python@v1         with:-          python-version: ${{ matrix.python-version }}+          python-version: ${{ matrix.py-version }}       - name: Build wheels-        if: github.event_name == 'push' || github.event_name == 'release' || matrix.python-version == env.OLDEST_PY_VERSION+        if: (matrix.os != 'ubuntu-18.04' || matrix.tf-version != '2.2.0rc1') && (github.event_name == 'push' || github.event_name == 'release' || matrix.py-version == env.OLDEST_PY_VERSION)

I prefer to err on the side of caution concerning the CI. Breaking the master branch means nobody can make new pull request and have the CI displaying meaningful results. I know it's very unlikely that we break it by using the 2.1.0 image. But there is very little benefit with working on this before the official image is released. I'd prefer to merge this and when the official image is released, we remove the exception in the yaml and fix potential incompatibilities. This new image will have python 3.8, will be possibly bigger and burst our disk size limit in github actions, I'd prefer we do it one step at a time.

gabrieldemarmiesse

comment created time in 2 days

Pull request review commenttensorflow/addons

Add tf 2.2.0rc1 in the build matrix.

 jobs:     strategy:       matrix:         os: ['macos-latest', 'windows-latest', 'ubuntu-18.04']-        python-version: ['3.5', '3.6', '3.7']+        py-version: ['3.5', '3.6', '3.7']+        tf-version: ['2.1.0', '2.2.0rc1']       fail-fast: false     steps:       - if: github.event_name == 'push'         run: echo "::set-env name=NIGHTLY_FLAG::--nightly"       - uses: actions/checkout@v2       - uses: actions/setup-python@v1         with:-          python-version: ${{ matrix.python-version }}+          python-version: ${{ matrix.py-version }}       - name: Build wheels-        if: github.event_name == 'push' || github.event_name == 'release' || matrix.python-version == env.OLDEST_PY_VERSION+        if: (matrix.os != 'ubuntu-18.04' || matrix.tf-version != '2.2.0rc1') && (github.event_name == 'push' || github.event_name == 'release' || matrix.py-version == env.OLDEST_PY_VERSION)

Because we don't have the docker image to build it yet :)

gabrieldemarmiesse

comment created time in 2 days

issue commenttensorflow/addons

TQDMProgressBar not working in TF-2.2.0rc1

Yeah we definitly lack some proper testing for the callbacks in Addons.

bjtho08

comment created time in 2 days

pull request commenttensorflow/addons

Gaussian Blur op

@ghosalsattam Maintainers are nearly all volunteers working on Addons during their free time, and the list of pull requests is quite long, it may take a few days, sometimes weeks to get a review depending on the size of the diff. No worries, as long as the build is passing and the pull request is opened, you'll get a review at some point.

ghosalsattam

comment created time in 2 days

push eventgabrieldemarmiesse/addons

gabrieldemarmiesse

commit sha e0e952a418e454b9f5ffb120bbdadf8eb20e7292

Remove info.

view details

push time in 2 days

push eventgabrieldemarmiesse/addons

Jhuo IH

commit sha fd5f43c57316dd78974ae26886822eb6af9551f4

Moved tests out of run_in_graph_and_eager_mode in multithread attendion (#1486) * Moved tests out of run_in_graph_and_eager_mode in multithread attendion * replace assert * solved conflict with #1436

view details

Jhuo IH

commit sha 603e7b52a53103e2951470fdce0fd9359d1ae23a

Moved tests out of run_in_graph_and_eager_mode in multithread attendion (#1487) * Moved tests out of run_in_graph_and_eager_mode in multithread attendion * cleanup

view details

Jhuo IH

commit sha f4933de4136562ca6222e93de7051b52977ea9ad

Moved tests out of run_in_graph_and_eager_mode in multithread attention (#1488)

view details

Jhuo IH

commit sha c0a65660f5d4c77056bd93cb055130fe22597014

Moved some tests out of run_in_graph_and_eager_mode in normalization (#1493)

view details

Gabriel de Marmiesse

commit sha 333bd29701c06b723327ab80ebb6aa7c8ec2caa8

Removed the automatic install of python packages in configure.py (#1462) * Removed the automatic install of tf. * No deps installed in the configure.py * Remove the no-deps. * Install in editable mode. * Hopefully a fix. * Use cache. * Fixing the gpu build. * Fix the gpu tests. * Fix install instructions. * Update contributing documentation * Update CONTRIBUTING.md Co-authored-by: Sean Morgan <seanmorgan91@gmail.com>

view details

gabrieldemarmiesse

commit sha 9b011c51dfea09490c168b37ca28c5a38f619787

Merge branch 'master' into matrix_tf220

view details

push time in 2 days

push eventgabrieldemarmiesse/addons

Jhuo IH

commit sha fd5f43c57316dd78974ae26886822eb6af9551f4

Moved tests out of run_in_graph_and_eager_mode in multithread attendion (#1486) * Moved tests out of run_in_graph_and_eager_mode in multithread attendion * replace assert * solved conflict with #1436

view details

Jhuo IH

commit sha 603e7b52a53103e2951470fdce0fd9359d1ae23a

Moved tests out of run_in_graph_and_eager_mode in multithread attendion (#1487) * Moved tests out of run_in_graph_and_eager_mode in multithread attendion * cleanup

view details

Jhuo IH

commit sha f4933de4136562ca6222e93de7051b52977ea9ad

Moved tests out of run_in_graph_and_eager_mode in multithread attention (#1488)

view details

Jhuo IH

commit sha c0a65660f5d4c77056bd93cb055130fe22597014

Moved some tests out of run_in_graph_and_eager_mode in normalization (#1493)

view details

gabrieldemarmiesse

commit sha 27ac6e805a778d79cc133dca656212ae57de877d

Merge branch 'master' into do_not_install_tf

view details

gabrieldemarmiesse

commit sha ec5a3674a7525c6029e258840dbf498b08ad32c0

Fix install instructions.

view details

push time in 2 days

push eventtensorflow/addons

Jhuo IH

commit sha c0a65660f5d4c77056bd93cb055130fe22597014

Moved some tests out of run_in_graph_and_eager_mode in normalization (#1493)

view details

push time in 2 days

PR merged tensorflow/addons

Moved some tests out of run_in_graph_and_eager_mode in normalization cla: yes layers

cc @gabrieldemarmiesse

+39 -35

1 comment

1 changed file

autoih

pr closed time in 2 days

more