| 6df1854646d9cc442ba109d3af15c2978d1599ac |
|
01-Nov-2017 |
Allen Lavoie <allenl@google.com> |
Add a way to get a list of variables from an optimizer.
optimizer.variables() returns a list of variables created by the optimizer in the current default graph. This is necessary in eager mode since collections are not practical and optimizer state needs to be checkpointed. Replaces tfe.get_optimizer_variables().
The implementation is basically collecting slot variables + Adam's two non-slot variables.
Alternatives considered:
- Collections for eager. We'd need weakref collections, since eager mode relies on variables being deleted when no longer in use. Weakref collections would lead to surprising behavior, since they would reflect currently existing variables rather than preserving them (variables could disappear between calls).
- A property rather than a method. Unfortunately this operation is rather expensive (the list needs to be constructed). A method better reflects that.
PiperOrigin-RevId: 174243995
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| a1d2a4ab90bd9df7312408f3971a2236810a1074 |
|
13-Feb-2017 |
Alexandre Passos <apassos@google.com> |
Backprop through resource handles.
Change: 147371282
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 67443722b26c3585d860d44e7069d997300a7187 |
|
31-Jan-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Enables sparse optimizers for resource variables.
Change: 146117209
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 58201a058853de647b37ddb0ccf63d89b2357f03 |
|
17-Dec-2016 |
Justine Tunney <jart@google.com> |
Remove hourglass imports from even more tests
Change: 142318245
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| c46f70c06985edce4905668a943801040c7a1354 |
|
16-Nov-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Makes the Variable.ref() method private.
Change: 139267342
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 2201ada5a00bb06148a9baf0d50c4dd07ecc4864 |
|
09-Nov-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Add support for sparse access to resource variables via embedding_lookup.
Change: 138642972
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| cbd3cacfb73bbea912b9d01c2540187684f751a7 |
|
03-Nov-2016 |
Illia Polosukhin <ipolosukhin@google.com> |
Replace usages initialize_all_variables -> global_variables_initializer
Change: 138128703
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 2da8f3106ab39c9971809fc67ca9c6eedfb98399 |
|
03-Nov-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Making simple gradient descent work with resource variables.
Change: 138121817
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 6132ae00e0914937d8416f37b969689148461d79 |
|
19-Sep-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Remove unnecessary imports in some python files.
Change: 133602571
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 0cf9ed3a719c0782695154d5a0bca260001cec15 |
|
02-Jun-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Update copyright for 3p/tf/python.
Change: 123900456
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 8cc567bf9703e14f0e16595eb9f220859a5eff20 |
|
26-May-2016 |
Vijay Vasudevan <vrv@google.com> |
Merge changes from github.
Change: 123342870
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 36c0475865ec103b5a3c2de4a69b17eddf7a9903 |
|
20-Apr-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Enable fp16 support for all optimizers, and also add unit tests for all that
have it (ie., everything except sync_replicas_optimizer). Mostly a matter of
adding the right casts for all the constant tensors, so that the clients
do not need to explicitly set types for them.
Also add a helper function assertAllCloseAccordingToType() that does comparison
with a larger epsilon for fp16 (adapted from training_ops_test.py and improved
somewhat).
Change: 120351448
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 7760ce56fc3ab4ab8cdc408e29d8ad8b539c417e |
|
11-Feb-2016 |
Josh Levenberg <josh11b@tensorflow.org> |
Get rid of some import cruft.
Change: 114374558
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| eecbcb261ad2a5d220c705dba6ff1e760adb8b95 |
|
27-Jan-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Changes Optimizer.compute_gradients to use var.ref() instead of var.value() as the tensor wrt which to compute the gradient. This enables things like this:
with tf.control_dependencies([barrier]):
# Fetch var just in time
v = tf.identity(var.ref())
# Now do something with v
Before this CL, the gradient wrt var would be 0 since tf.identity(var.ref()) does not depend on var.value() (which is another tf.identity(var.ref()), but from which there is no path in the graph to the loss).
gradient_descent_test.py tests the behavior fixed by this CL.
Change: 113163441
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 795f35da2d458cbae477ac2fe2bff80c1427a771 |
|
01-Dec-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: upstream changes to git
Change:
Clean up documentation for ReverseSequence
Change:
Updated several tensorflow operations to use 32bit indices on GPU.
Change:
Add attribute batch_dim to ReverseSequenceOp.
Change:
Fix error in convert_to_records.py. As reported in
https://github.com/tensorflow/tensorflow/issues/370
by AlexUnderMicrocontRoll.
Change:
Update TensorBoard README.
Change:
Fixes to boolean flags reported in
https://github.com/tensorflow/tensorflow/issues/379. Supports:
--bool_flag=True --> True
--bool_flag=False --> False
--bool_flag=gibberish --> False
--bool_flag --> True
--nobool_flag --> False
Fixes #379
Change:
Update generated Op docs.
Change:
Enable local development of TensorBoard using gulp
Also make tf-tensorboard a regular component rather than special case
This is mostly effected by creating tfserve.js, which is a small server
with clever routing to load from bower_components/ and components/ using
the paths that work within google3.
Workflow: `gulp serve`
Change:
Add a full working code example to the tensorboard and summaries tutorial
Change:
Fix seq2seq_test when running on GPU.
The "proj_w" and "proj_b" variables were being created before the
`test_session()`'s device function took effect, which pushed the
placement algorithm into making an incorrect decision.
Change:
Add a sentence in TensorBoard README on how to serialize summary data to logs and provide link to the how-to tutorial on the TensorFlow website.
Change:
Add error-catching code if string_input_producer is supplied a null input.
Before this change, it would die with an opaque shape error from inside
the queue. This change catches (most) python null lists being
passed directly in, and at runtime detects null tensors.
Adds two tests for this to input_test.py
Change:
Speed up for models that use the same variable multiple times in the case
where variables must be copied across devices:
- Have Variables wrap the Variable op in an Identity op when converted to Tensor.
This avoids multiple copies across devices if a variable is used multiple time
in a computation.
- Add Variable.mutable() to return the non-wrapped Variable op for used when
assigning new values.
- Add an as_ref parameter to convert_to_tensor() to allow code to specify
if they plan to assign a new value to the result of the conversion. Make Variable
return the result of Variable.mutable() when as_ref is True.
- Make all ops that assign values to variables pass as_ref=True when converting
their arguments.
Change:
Change to reduce critical section times in gpu_event_mgr.h:
(1) Call stream->ThenRecordEvent outside the EventMgr critical section
(2) Do memory deallocation outside the critical section
Speeds up one configuration of ptb_word_lm from 2924 words per
second (wps) to 3278 wps on my desktop machine with a Titan X.
Change:
Remove some colons that break the open source build
::tensorflow::StringPiece breaks for @raingo, see
https://github.com/tensorflow/tensorflow/issues/358.
tensorflow::StringPiece (without the leading colons)
seems to fix the problem.
Change:
Added check that inputs to Operation is a list and make a defensive copy of the input. This is for cases where the input list is changed such as in _add_input.
Change:
Use standard names for TensorFlow dtypes in the tutorial.
Change:
Add tests for tensor inputs.
Change:
Fix build after declaring more types for ops
Change:
Switch to 32 bit indexing to speedup convolutions and concatenations.
Change:
Add convert_image op to convert between types for images (similar to OpenCV's cvtScale).
Change:
Make cast work between numeric types (bool, uint8, int16, int32, int64, float, double).
Change:
Padding input data for odd number of paddings, so we can use cudnn anyway.
+ Fix total padding computation when padding==VALID.
+ This CL makes the Googlenet benchmark run 5x faster.
Change:
Support IndexedSlices in ConcatGrad
Change:
* sampled softmax op uses one embedding lookup for positive and negative samples
* float64 support for sampled softmax
Change:
Move RNN code out of models.rnn (without breaking existing code). The API may still undergo minor changes, until full documentation as added.
Change:
Changed to use per-step stacks for the accumulators used in while-loop gradient computation. This addresses the problem caused by using concat without sufficient static shape information. It should also improve performance as we avoided those expensive concats.
Change:
Update generated Op docs.
Change:
Improve error messages when the optimizer finds no variables to minimize or
when none of the variables has gradients.
Change:
Say that -1 isn't just for flattening in reshape docs
Also add scalar reshape (reshape(t, [])) as an example.
This fixes https://github.com/tensorflow/tensorflow/issues/281.
Change:
This is a test.
Base CL: 109118714
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| 9c3043ff3bf31a6a81810b4ce9e87ef936f1f529 |
|
20-Nov-2015 |
Manjunath Kudlur <keveman@gmail.com> |
TensorFlow: Improve performance of Alexnet
Changes:
* error message that refers to removed `DefaultSession` method.
* -Wnull-conversion warnings
* the "_start_time" attr for recvs when the flag "--brain_enable_scheduling_for_recvs" is set.
* typo in tutorial data download progress message.
* a typo ("however their installing"=>"however installing").
* typo, rename "TensorFlow Mechanics" to "How To" to be consistent with the website.
* a typo ("subtact"=>"subtract").
* protobuf examples in comments in tensorflow::Example.proto.
* formula formatting in MNIST beginner tutorial
* negative fraction-of-queue-full stats
* protobuf inclusion path so that Android demo will build under Blaze.
* small typo (moderatly > moderately)
* Session.run() to check that tensor arguments come from the session's graph.
* another six import
* seq2seq typo in bazel command
Base CL: 108349164
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| f2102f4e2c1c87f1d1bf9ab856a2849c54478760 |
|
12-Nov-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: upstream changes from the afternoon.
Changes:
- futurize --stage2 changes for Python 3 compatibility by @girving.
- Small updates to documentation by @vrv, schuster and others
- Account for failure of std::thread::hardware_concurrency by @ebrevdo.
- More changes for backwards-compatibility tests by Josh
- Updates to python op doc generation by Josh
- Added support for using the best-fit allocator via ConfigProto by @vrv.
- Rename LocalSession to DirectSession, since local was a bad name for
it.
- Enable tf.nn.moments() to work with tensors of unknown shape by @mrry.
GITHUB_ISSUE: 139
- Changes for Android build by Andrew.
Base CL: 107645181
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|
| f41959ccb2d9d4c722fe8fc3351401d53bcf4900 |
|
07-Nov-2015 |
Manjunath Kudlur <keveman@gmail.com> |
TensorFlow: Initial commit of TensorFlow library.
TensorFlow is an open source software library for numerical computation
using data flow graphs.
Base CL: 107276108
/external/tensorflow/tensorflow/python/training/gradient_descent_test.py
|