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

Branch 199474340 #19816

Merged
merged 42 commits into from
Jun 6, 2018
Merged

Branch 199474340 #19816

merged 42 commits into from
Jun 6, 2018

Conversation

drpngx
Copy link
Contributor

@drpngx drpngx commented Jun 6, 2018

No description provided.

shashishekhar and others added 30 commits June 5, 2018 10:22
@shashishekhar @tensorflower-gardener
Internal change.
c8090fa 
PiperOrigin-RevId: 199316885
@tensorflower-gardener
Respect name scopes opened in tower mode when creating vars in cross …
a7c026e 
…tower mode.

PiperOrigin-RevId: 199319758
@tensorflower-gardener
Do not enable tensor ops for cuDNN RNN unless explicitly specified.
b2e5670 
PiperOrigin-RevId: 199321021
@caisq @tensorflower-gardener
Fix bug in which uncompiled tf.keras.Models cannot be saved
e86d969 
This bug seems to be specific to tf.keras, i.e., it doesn't happen to keras.

PiperOrigin-RevId: 199334073
@tensorflower-gardener
Add core/util/exec_on_stall.h a tool for debugging deadlocks with
b1fd2ef 
less logging.

PiperOrigin-RevId: 199334548
@tensorflower-gardener
Extend and refactor reader_ops_test
62a70dd 
PiperOrigin-RevId: 199335030
@tensorflower-gardener
Implementation of the symmetrically quantized LSTM TFLite Op.
920df27 
PiperOrigin-RevId: 199337082
@ezhulenev @tensorflower-gardener
Move ReplaceMulWithSquare to a separate optimizer stage.
2b5f598 
PiperOrigin-RevId: 199338297
@tensorflower-gardener
Fixing typo in Subtract Kernel.
a1e2587 
PiperOrigin-RevId: 199340127
@tensorflower-gardener
Clarifies how to pass training hooks to TPUEstimator in the docstring…
b7928ac 
… for TPUEstimator.

PiperOrigin-RevId: 199341721
@ezhulenev @tensorflower-gardener
Move SimplifyAggregation to separate aggregation stage.
c681be0 
PiperOrigin-RevId: 199346067
@skye @tensorflower-gardener
Introduce tf.contrib.control_flow.new_cond.
1bac618 
new_cond is a new implementation of tf.cond. Instead of emitting
control flow ops (i.e. Switch and Merge nodes), new_cond emits a
single If op, which represents the conditional branches as TF
functions.

With this change, users can use new_cond and take its gradient.

The idea is for new_cond to eventually replace tf.cond. There are
several functional and performance gaps that must be addressed first,
including:
* Gradients won't work on imported graphs
* Misc. limitations of TF functions (lack of collections, device scopes, etc.)
PiperOrigin-RevId: 199346735
@tensorflower-gardener
Update ops-related pbtxt files.
1e92632 
PiperOrigin-RevId: 199347316
@tensorflower-gardener
Allow calling getanno with a default value. Failure is still the defa…
70a96b5 
…ult behavior.

PiperOrigin-RevId: 199349592
@bixia1 @tensorflower-gardener
Fix test MultiOutputReduceFusionScalar to use an identity value as re…
92ceec1 
…duction init_value.

PiperOrigin-RevId: 199350818
@tensorflower-gardener
Go: Update generated wrapper functions for TensorFlow ops.
c03d2c4 
PiperOrigin-RevId: 199351707
@tensorflower-gardener
No longer assume that the default job is "localhost" in graph mode
12b20a5 
DistributionStrategy, since it depends on the session. Drop
"job:localhost" when canonicalizing in graph mode.

PiperOrigin-RevId: 199354215
@tensorflower-gardener
Update TOCO Python command line flags.
d935dd9 
PiperOrigin-RevId: 199361276
@tensorflower-gardener
[XLA] Redesign: delete SessionModule.
f023073 
PiperOrigin-RevId: 199361402
@tensorflower-gardener
Typo fix in suggested pip message for tpu cluster resolver.
e0c9871 
PiperOrigin-RevId: 199362908
@tensorflower-gardener
Correctly implement the checks for getanno.
7638924 
PiperOrigin-RevId: 199366963
@ezhulenev @tensorflower-gardener
Move fold-transpose and fold-conjugate optimizations into stages.
5c1c4fc 
PiperOrigin-RevId: 199371452
@yifeif @tensorflower-gardener
internal change
e545463 
PiperOrigin-RevId: 199372205
@skye @tensorflower-gardener
Disable flaky test for now.
73c3a8a 
PiperOrigin-RevId: 199373124
@tensorflower-gardener
Only calls compare function if values were read from event file
ece5f51 
PiperOrigin-RevId: 199373169
@tensorflower-gardener
Updates Python TOCO command line and TOCO documentation.
677c83e 
PiperOrigin-RevId: 199375811
@saxenasaurabh @tensorflower-gardener
Support uint8, int32 and int64 for SpaceToDepth in TOCO.
135a259 
PiperOrigin-RevId: 199376731
@tensorflower-gardener
[XLA] Make CrossReplicaSum support general cross replica reduce. Also…
a57f0de 
… change the interface to be able to describe the common AllReduce semantic.

PiperOrigin-RevId: 199376926
@tensorflower-gardener
Add the dart rule to tensorflow/core:protos_all.
902832a 
PiperOrigin-RevId: 199377753
@tensorflower-gardener
[XLA] Add a bytes read+written table to the end of --xla_hlo_profile.
8a14185 
This is useful when tuning fusion heuristics -- you expect this number
to go down (even if the total runtime doesn't go down, due to suboptimal
emitters).

PiperOrigin-RevId: 199386923
Michael Case and others added 12 commits June 5, 2018 17:50
@tensorflower-gardener
Moves generated android_sdk() and android_ndk() repo rules out of WOR…
5105350 
…KSPACE.

These rules currently get written by configure.py script to WORKSPACE
file which is not ideal since (1) WORKSPACE file is tracked by git and
(2) we require users to manually delete the rules in order to
update/regenerate them.

Moving these rules into an external repo that is generated based on
several ENV variables set by the configure.py script. Modifying any
of these ENV variables will cause the rules to be updated.

PiperOrigin-RevId: 199388460
@tensorflower-gardener
Automated g4 rollback of changelist 197562826
2366bd0 
PiperOrigin-RevId: 199388675
@shashishekhar @tensorflower-gardener
Fix iOS build.
acded19 
PiperOrigin-RevId: 199395164
@tensorflower-gardener
Add more logging to report module group metadata statistics.
98be57e 
PiperOrigin-RevId: 199397890
@tensorflower-gardener
Add __init__.py to all_reduce.
0978455 
PiperOrigin-RevId: 199399375
@gunan @tensorflower-gardener
Disable testLargeCase in metric_ops_test
95cd2d4 
PiperOrigin-RevId: 199405764
@akshayka @tensorflower-gardener
Minor touch ups to PartitionedCallOp.
76c9358 
Mostly just cosmetic refactoring to make PartitionedCallOp more readable; also registers a GPU kernel.

PiperOrigin-RevId: 199433460
@tensorflower-gardener
Tensorflow protos allow enum values outside of the listed constants; …
abd8348 
…this is now properly supported in the Text Format.

PiperOrigin-RevId: 199450074
@tensorflower-gardener
Internal change
47d42a1 
PiperOrigin-RevId: 199464493
@alextp @tensorflower-gardener
Error checking in c/python code.
c1b9ac9 
PiperOrigin-RevId: 199465056
@jtkeeling @tensorflower-gardener
Clarify documentation of Dataset.filter
5c26ec2 
It was not explicitly stated that the predicate should return True for elements the user wants to keep.

PiperOrigin-RevId: 199474340
Merge commit for internal changes
b458040
@drpngx drpngx requested a review from yifeif June 6, 2018 16:44
@drpngx drpngx added the kokoro:force-run Tests on submitted change label Jun 6, 2018
@kokoro-team kokoro-team removed the kokoro:force-run Tests on submitted change label Jun 6, 2018
@drpngx
Copy link
Contributor Author

drpngx commented Jun 6, 2018

Timeout, retrying (dataset_optimize_op_test).

@drpngx drpngx merged commit bef709a into tensorflow:master Jun 6, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@yifeif yifeif yifeif approved these changes

Assignees
No one assigned
Labels
cla: yes
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet