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

Upgrade to newer tensorflow #3

Draft
wants to merge 214 commits into
base: master
Choose a base branch
from
Draft

Upgrade to newer tensorflow #3

wants to merge 214 commits into from

Conversation

erikmchut
Copy link

Upgrade to newer tensorflow

alextp and others added 30 commits November 11, 2019 09:46
@alextp
Merge pull request tensorflow#34160 from Intel-tensorflow/eager_test_…
1ce5658 
…compilation_fix

[r2.1-CherryPick]: Fixing compilation error in MKL eager test
@geetachavan1
Re-add the supported CUDA compute capabilities into the toolchain cre…
f785554 
…ated for manylinux2010 compatibility for the 10.1 toolchain.

PiperOrigin-RevId: 279422340
Change-Id: Icaff93aed22c3a319efb6dee4c3f703eb17c1a11
@geetachavan1
Add a missing line for compute in toolchain.
236128a 
PiperOrigin-RevId: 279460905
Change-Id: Iaf895d72256b826aa2d3e474033800ececc7f658
@tensorflower-gardener @geetachavan1
Export Keras preprocessing layers in tf.keras.layers.experimental.
d29ff10 
PiperOrigin-RevId: 279470036
Change-Id: Ib2b363d21baa14ac49088dd700f769c933164340
@geetachavan1
run Model.*_on_batch using tf.function where appropriate.
f8759b6 
PiperOrigin-RevId: 279412139
Change-Id: I3c55a5388b0cfd144d63f5e9f444cf3e96471ec4
@aaudiber @geetachavan1
Fix flaky doctest.
b22b9c9 
The doctest was using dictionary output to exhibit structured
element usage. This caused a problem since dictionary key order
is undefined. This changes the example to use a tuple instead.

PiperOrigin-RevId: 279359482
Change-Id: Ideb9970935079b156f88f48672dea2d375d497ae
@goldiegadde
Merge pull request tensorflow#34170 from geetachavan1/cherrypicks_6VUBZ
f455bcf 
[r2.1:Cherrypick] Add a missing line for compute in toolchain and Re-add the supported CUDA compute capabilities into the toolchain
@goldiegadde
Merge pull request tensorflow#34172 from geetachavan1/cherrypicks_5KWST
0ccee2f 
[r2.1:Cherrypick] Export Keras preprocessing layers in tf.keras.layers.experimental.
@goldiegadde
Merge pull request tensorflow#34173 from geetachavan1/cherrypicks_TK4O4
89487ef 
[r2.1:Cherrypick] Run Model.*_on_batch using tf.function where appropriate.
@aaudiber
Fix bug in output ordering from interleave with num_parallel_calls !=…
e8c1190 
… None.

Before this fix, Dataset.interleave(..., num_parallel_calls=2) would result
in non-deterministic output order, even when
options.experimental_deterministic=True. The non-determinism is seen
when the cycle length of the interleave exceeds the number of input elements.

PiperOrigin-RevId: 279838573
Change-Id: I6dc94b071a8eeb269e24b192aa65be105022444c
@goldiegadde
Merge pull request tensorflow#34175 from geetachavan1/cherrypicks_ZLL35
9411012 
[r2.1:Cherrypick] Fix flaky doctest.
@goldiegadde
Merge pull request tensorflow#34180 from aaudiber/cherrypicks_7EUZ3
1a0c306 
Fix bug in output ordering from interleave with num_parallel_calls
@rxsang
Add XLA context to the function cache key
5b2a796
@alextp
Merge pull request tensorflow#34208 from tensorflow/rxsang-patch-1
b5d0f01 
Add XLA context to the function cache key
@jsimsa
[tf.data] Improving detection of infinitely repeated datasets in the …
7ead42a 
…presence of errors.

PiperOrigin-RevId: 280200721
Change-Id: Icfaffb567b970da140e9b0d3a6c2093452893f01
@jalexstark @jdduke
Ruy: Optimize (partial) of x86 AVX-512 8-bit pack.
664193e 
PiperOrigin-RevId: 280241069
Change-Id: I6cafe2b0e7e2dc13c098fd80780d9aa79d2a217b
@jalexstark @jdduke
Ruy: Further optimization (partial) of x86 AVX2 8-bit pack.
1c52525 
PiperOrigin-RevId: 280241172
Change-Id: I7e641a9ff00086f67e5e93676764799d44ef96d6
@tensorflow-jenkins
Insert release notes place-fill
3f150ee
@tensorflow-jenkins
Update version numbers to 2.1.0-rc0
0738369
@karimnosseir
Label libs with alwayslink
0bd44b7 
PiperOrigin-RevId: 279344508
Change-Id: Id43dcf4dafc8d45eb4eacd5f2730a6fad0ddf133
@tensorflower-gardener @karimnosseir
Add more alwayslink=1 to cc_library as a follow up on flipping the …
422496e 
…default of --incompatible_remove_legacy_whole_archive in cl/277339372

This should fix the latest issue reported in tensorflow#33415.
Also fixes an internally reported missing symbol.

Related to bazelbuild/bazel#7362

PiperOrigin-RevId: 279792794
Change-Id: I6f5d26ee37b9c886662df5e2daf9273c15cae865
@karimnosseir
Merge pull request tensorflow#34252 from karimnosseir/cherrypicks_R63JV
512ccef 
Cherrypicks r63 jv
@jsimsa
[tf.data] Improving detection of infinitely repeated datasets in the …
f333bd6 
…presence of errors.

PiperOrigin-RevId: 280200721
Change-Id: Icfaffb567b970da140e9b0d3a6c2093452893f01
@jsimsa
[tf.data] Improvements to cancellation logic.
a594b6b 
This CL:
- makes sure that a previously cancelled input pipeline won't hang in ParallelMapDataset kernel
- excludes errors::Cancelled() from being ignored by the IgnoreErrorsDataset kernel to avoid infinitely looping after a cancellation

PiperOrigin-RevId: 280425330
Change-Id: If210c94c59e3b33b170e77aea6c87308241364e6
@goldiegadde
Merge pull request tensorflow#34241 from jsimsa/cherrypicks_H0PP4
8959ed1 
2.1-RC0 cherry-pick request: [tf.data] Improving detection of infinitely repeated datasets in the presence of errors.
@jsimsa
Merge branch 'r2.1' into cherrypicks_P96RK
4ba2df6
@jaingaurav
Support LogicalDevice in MirroredStrategy config
eb18ad3 
PiperOrigin-RevId: 280290757
Change-Id: I52dfff634e6e0ccdc81cd5cce682d7df3499b618
@geetachavan1
Update BUILD
bc75348
@rohan100jain @jsimsa
OwnedIterators and MultiDeviceIterators get created using their compo…
e16b7cc 
…nents at least twice during function tracing. We don't want to run the deleter in those cases as they are just referring to the original resource and not creating a new one. So we just create the deleter the first time.

PiperOrigin-RevId: 280071920
Change-Id: I266ac500246354e6e4d6145835c4f90926ec5715
@geetachavan1
Update BUILD
a5c0ac6
goldiegadde and others added 28 commits December 20, 2019 16:35
@goldiegadde
Merge pull request tensorflow#35041 from reedwm/r2.1
87143b5 
[r2.1 Cherrypick]: Unexpose LossScaleGradientTape.
@mihaimaruseac
Typo from solving merge conflict
bd74d92
@goldiegadde
Update RELEASE.md
5dc7e36
@goldiegadde
Merge pull request tensorflow#35273 from tensorflow-jenkins/version-n…
3f078b7 
…umbers-2.1.0rc2-32619

Update version numbers for TensorFlow 2.1.0-rc2
@goldiegadde
Update RELEASE.md
b8132f4
@goldiegadde
Merge pull request tensorflow#35322 from tensorflow/mm-cp-segfault-fix
7391285 
Fix segfault when attempting to convert string to float16.
@goldiegadde
Merge pull request tensorflow#35324 from tensorflow/goldiegadde-patch-1
c23091d 
Update RELEASE.md
@goldiegadde
Merge pull request tensorflow#35238 from ROCmSoftwarePlatform/r2.1-ei…
9837ece 
…gen-patch

[ROCm][r2.1] eigen patch needed for HIP header changes
@mihaimaruseac
Fix bazelisk link
eaa4090
@bmzhao
Fixing .bazelversion to be the same version that CI uses for 2.1.0: h…
f61b498 
…ttps://github.com/tensorflow/tensorflow/blob/v2.1.0-rc2/tensorflow/tools/ci_build/install/install_bazel.sh#L18
@mihaimaruseac
Cherry-pick the release build scripts
5a3c577
@mihaimaruseac
Merge pull request tensorflow#35546 from tensorflow/mm-r21-presubmit-…
32095d6 
…scripts

Cherry-pick the release build scripts
@mihaimaruseac
Merge pull request tensorflow#35470 from bmzhao/bazelversion-cherrypick
66191e1 
Update .bazelversion
@Artem-B @compnerd
Provide integer variants of isnan/isfinite to work around a quirk in …
2d5cbf1 
…MSVC standard library.

PiperOrigin-RevId: 286447750
Change-Id: I5af8054ac69782e351cff440d88824eade90f963
@tensorflower-gardener @compnerd
Remove designated initializers.
4dfee0d 
PiperOrigin-RevId: 286436468
Change-Id: I61c4337f54f3a2533a03de65371c2aa1b1a76e5b
@yunxing @compnerd
[XLA] Remove use of designated initializers in dynamic dimension infe…
4c07219 
…rence.

PiperOrigin-RevId: 286330416
Change-Id: I5878ab23b9596ac0c5dbf91394ee6ba8e5683c77
@Artem-B @compnerd
Define _USE_MATH_DEFINES for windows builds.
9badef2 
PiperOrigin-RevId: 286642121
Change-Id: I8ec228fd768df5b20a11cdc63fcf219d6ab24de7
@compnerd
configure.py: add -D_USE_MATH_DEFINES manually
a8adad9 
This adds `--ccopt -D_USE_MATH_DEFINES` manually to the .bazelrc as the
verison of bazel used on r2.1 (0.29.x) does not support platform
specific options.  This allows the windows build to function by default.
@mihaimaruseac
Add Python3.7 testing on MacOS as we drop support for Python2.
5b7addf 
To be cherry-picked on `r1.15`, `r2.0`, and `r2.1` branches.

PiperOrigin-RevId: 287871757
Change-Id: Ic530e884de421a39a82c686f1f0d086b6400d75c
@mahmoud-abuzaina
Cherry-picking 55e20a6 - klockwork fix
3ed1f02
@mihaimaruseac
Merge pull request tensorflow#35566 from tensorflow/mm-r21-macos-py3
5efe38d 
Add Python3.7 testing on MacOS as we drop support for Python2.
@mihaimaruseac
Merge pull request tensorflow#35567 from Intel-tensorflow/tf-2.1-kloc…
8427475 
…work-cherry-pick

[r2.1:Cherrypick][Intel MKL] Cherry-picking 55e20a6 - Klockwork fix
@mihaimaruseac
Merge pull request tensorflow#35548 from compnerd/r2.1-windows-build
bd56e04 
r2.1 cherry-pick request: windows build support
@tensorflow-jenkins
Update version numbers to 2.1.0
d43e3c7
@goldiegadde
Merge pull request tensorflow#35620 from tensorflow-jenkins/version-n…
e5bf8de 
…umbers-2.1.0-23859

Update version numbers for TensorFlow 2.1.0
@mralbean
Update c api visibility
9f67747
@erikmchut
Silence build warnings in tflite
26dfffb 
Silence build warnings in tflite
Upgrade to newer tensorflow
db4dfbd
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.