| 4861f6128d4cc30583c2f5370df4e021e8045af6 |
|
16-Feb-2018 |
Martin Wicke <577277+martinwicke@users.noreply.github.com> |
Merge pull request #16085 from yongtang/16034-dropout-noise_shape
Add unspecified dimensions (-1) support for noise_shape with tf.nn.dropout
|
| ad07a86d75ab06bbcfd6f8f6a24debd9036a52d0 |
|
24-Jan-2018 |
Jianwei Xie <xiejw@google.com> |
Fixed linter errors.
PiperOrigin-RevId: 183115307
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| d9f93c42a50b1f1401d9c186eac0ae8dc9093c3b |
|
24-Jan-2018 |
Jianwei Xie <xiejw@google.com> |
Merge changes from github.
PiperOrigin-RevId: 183100142
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| ae5e7ffeb7f5144f66350624c75fe3dfd0ecd8ce |
|
24-Jan-2018 |
Yong Tang <yong.tang.github@outlook.com> |
Use None instead of -1 for partial shape
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 63db01abde2ad4977587591b1646f40b175f7faa |
|
12-Jan-2018 |
Yong Tang <yong.tang.github@outlook.com> |
Add test case for unspecified dimensions support of noise_shape in tf.nn.dropout
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| ec804af7d84034868701b428a9f28bf6dd98deb1 |
|
12-Jan-2018 |
Olivia Nordquist <nolivia@google.com> |
fixes _softmax function so that nn.softmax(...) and nn.log_softmax(...) can accept negative axes that aren't -1 as the documentation claims. _swap() was accepting negative indices when math_ops.range() does not accept negative indices
PiperOrigin-RevId: 181771157
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 47249f349d13f5a11a8dc8c4026c54b49c88cfe0 |
|
20-Dec-2017 |
Yao Zhang <yaozhang@google.com> |
Extend DataFormatDimMap to handle tensors.
PiperOrigin-RevId: 179726269
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 9383f0b4dd7d06cd4a54f8e4df07182a90530690 |
|
16-Dec-2017 |
Yao Zhang <yaozhang@google.com> |
First and second dimension should be reversed.
PiperOrigin-RevId: 179308396
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| d52113a3b0bb3cc5e52e8fdbc0b2c98675b37d8c |
|
15-Dec-2017 |
Yao Zhang <yaozhang@google.com> |
Support permute 2d tensors.
PiperOrigin-RevId: 179229453
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| a94506b831aac707979aa615b0f424f2d4b2f316 |
|
13-Dec-2017 |
Yao Zhang <yaozhang@google.com> |
Support permutation from NCHW to NHWC.
PiperOrigin-RevId: 178824999
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| d94d6c45a56bc0871bb152a18818a1def3e17a7d |
|
08-Dec-2017 |
Yao Zhang <yaozhang@google.com> |
Add DataFormatVecPermute op.
PiperOrigin-RevId: 178421287
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 63dd8ffea012bf7c743d6848faf5f406ede94c05 |
|
06-Dec-2017 |
Yao Zhang <yaozhang@google.com> |
Add DataFormatDimMap op.
PiperOrigin-RevId: 178051608
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| cadcda216ec7d6f5f3e36dfc7863634f4f03f71f |
|
18-Oct-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Implementation of the Swish activation function.
PiperOrigin-RevId: 172651500
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 7a8c63da365106048dc96affddb39e2fdc33da89 |
|
12-Sep-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Migrate `leaky_relu` to `nn_ops.py`. Will be used for TFGAN.
PiperOrigin-RevId: 168386268
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| b6409594d39954c1794d14f6d4f6d9a5b40125b3 |
|
19-Aug-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Convert some tests to cover both eager and graph.
PiperOrigin-RevId: 165760364
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| fde5f92acfb5a6ef0ac21cfe6a420785f6e3ca24 |
|
15-Aug-2017 |
Igor Saprykin <isaprykin@google.com> |
Clarify that `x` in `tf.nn.dropout` needs to be a floating point tensor.
`x` is going to be scaled up, so it doesn't make sense for it to be of another type.
PiperOrigin-RevId: 165348557
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| fbe6ae13a36c80454f032a37577ab199a67ef01d |
|
25-Jul-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Add gradient checking tests for nn.moments().
PiperOrigin-RevId: 163110994
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| eccd162119675d0bf5bc6f8e6a93dcda7ab6db4a |
|
05-Jul-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Fix moments calculation to never result in negative variance and avoid
doing extra work when shift = None. With the current calculation shift is
ignored.
PiperOrigin-RevId: 161003939
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| cd15cb8206d19dde403e9e9c2293fa9dea7ec9ee |
|
30-Jun-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Remove unused internal facility for loss functions to colocate logit computations with the weights.
PiperOrigin-RevId: 160592547
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| a1e054fea7861a1111e0df8b83945607a171b7f4 |
|
22-Jun-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
For candidate sampling, add facility to colocate the logit computation with the sharded embeddings.
PiperOrigin-RevId: 159854706
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| a99694277e33fe8e64fb1a87522704f263a6fb5a |
|
10-Apr-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Support partitioned bias variable in sampled_softmax_loss.
Change: 152730499
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| ab98b2e4c378e62e9d7a4fbc1fda090083db0bcc |
|
15-Feb-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Create test that validates relu(nan) = nan
Change: 147527599
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| fb53109ad20b2e568246b05796296a047c271856 |
|
26-Jan-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
A fix for a softmax failure when dim is explicitly set to the last dim.
Change: 145631616
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| fdbd02c8d7f07bd1207938662716fad8857dcd55 |
|
10-Jan-2017 |
Sergio Guadarrama <sguada@google.com> |
Make moments numerically stable by default. Added tests for moments.
Change: 144114955
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 58201a058853de647b37ddb0ccf63d89b2357f03 |
|
17-Dec-2016 |
Justine Tunney <jart@google.com> |
Remove hourglass imports from even more tests
Change: 142318245
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| a46775975ded8b6f8935d996d48872ba319a9733 |
|
13-Dec-2016 |
Jianwei Xie <xiejw@google.com> |
Swap the order of inputs and labels of the sampled_softmax_loss.
Change: 141900201
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 761b12ed82d31195de002cdd687cbb77d7aba628 |
|
07-Dec-2016 |
Jianwei Xie <xiejw@google.com> |
Swap the input and label arguments in nce_loss
Change: 141244045
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| ee367604be41d9326f082d911efc4b7596f95ac3 |
|
02-Dec-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Swaps the predictions/labels arguments in log_poisson_loss.
Change: 140884027
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| af6c3f0cd4d35233e25a7c8a3994eaca412b7996 |
|
23-Nov-2016 |
Patrick Nguyen <drpng@google.com> |
Seal tf.nn's interface.
This moves functions that were defined directly in nn.py
into nn_ops.py.
Change: 139979007
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 8945cecc1e78f9d9647eb93f3a2bda0b994d4bb3 |
|
24-Aug-2016 |
Vincent Vanhoucke <vanhoucke@google.com> |
Implement the Concatenated ReLU (CReLu) non-linearity. This non-linearity makes it possible to reduce the depth of lower layers in a convolution tower by a factor of two at no cost in accuracy. See: http://arxiv.org/abs/1603.05201
Change: 131114642
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 3b7153de39732170cc5abd01b0e051bf98e066eb |
|
22-Aug-2016 |
Vijay Vasudevan <vrv@google.com> |
Fix L2Normalize when passing a list of dims. Fixes #3932.
Change: 130947456
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 542de3163a95d1487c6d9f4b7f8570571916ff23 |
|
22-Aug-2016 |
Geoffrey Irving <geoffreyi@google.com> |
Make dropout avoid work if keep_prob == 1
This speeds up models which sometimes use dropout and sometimes don't, without complicating them in place.
Change: 130903091
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 6ab68004062d35b1c668c7d743cc5e8296edd825 |
|
28-Jul-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Adding log poisson loss in order to support Poisson Regression.
Change: 128637765
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| b1b2dc893d616c024c5390dae8b2f932c917d7f8 |
|
07-Jun-2016 |
Martin Wicke <wicke@google.com> |
Merge changes from github.
Change: 124183870
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 0cf9ed3a719c0782695154d5a0bca260001cec15 |
|
02-Jun-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Update copyright for 3p/tf/python.
Change: 123900456
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 1d094c93bfde6f8df458f15527ea377e0b92e639 |
|
13-May-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Add gradient for log_softmax and test.
Change: 122290048
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| d107fecfb47d32777708025449fc6a2527ea1fcf |
|
12-May-2016 |
Martin Wicke <wicke@google.com> |
Split nn_test into more manageable pieces.
Change: 122163424
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| adcee0e21085f28d79ad3a021c0be481c63c5a10 |
|
10-May-2016 |
David Soergel <dev@davidsoergel.com> |
This change causes Ops with multiple outputs to return them as a NamedTuple
instead of as a plain Tuple, allowing Python callers to distinguish the outputs
by name. This is fully backwards-compatible, as a NamedTuple can be unpacked
just like a regular tuple.
Example code generated in gen_array_ops.py:
_unique_with_counts_outputs = ["y", "idx", "count"]
_UniqueWithCountsOutput = collections.namedtuple("UniqueWithCounts",
_unique_with_counts_outputs)
def unique_with_counts(x, name=None):
...
result = _op_def_lib.apply_op("UniqueWithCounts", x=x, name=name)
return _UniqueWithCountsOutput._make(result)
Change: 121965295
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| a1bc10f9e28cd3d33e91bfaedd8199e4590f2893 |
|
04-May-2016 |
Gunhan Gulsoy <gunan@google.com> |
Fix nn_test.py
Change: 121486952
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 7ae67a14e273a000e5b306d335dc0f1d7c0650c1 |
|
04-May-2016 |
Benoit Steiner <benoit.steiner.goog@gmail.com> |
Added support for fp16 to the l2loss kernel
Change: 121447446
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| c1aa6423e35803d08b7092a874632b6b7b13a130 |
|
21-Apr-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Add atrous convolution (atrous_conv2d op). Fixes #1815.
Change: 120462895
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| edb0ad88c99616635c5dbc921d6a5fb152ee6dab |
|
31-Mar-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Adds weighted_cross_entropy that allows one to trade off recall vs. precision penalty.
Change: 118624065
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 8016e72bae635392b06598b74c6f87d452e07034 |
|
24-Mar-2016 |
Olivia Nordquist <nolivia@umich.edu> |
fixing linting errors
Change: 118049024
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 0bcaf2781eace3731c0ac07dea82920b07b6c2a4 |
|
14-Mar-2016 |
Vincent Vanhoucke <vanhoucke@google.com> |
Another 1 in 1000 numerical failure.
Change: 117155975
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| bc27f89b872153d0517a7881da13124f56b8c2ab |
|
14-Mar-2016 |
Vincent Vanhoucke <vanhoucke@google.com> |
De-flakify testBatchNormArbitraryShapes
Change: 117146617
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 80ddfbf11f63cf5d55ad16f4bc871cdd18ced3da |
|
02-Mar-2016 |
Derek Murray <mrry@google.com> |
Add shape verification for `tf.nn.sigmoid_cross_entropy_with_logits()`.
Fixes #1343.
Change: 116175143
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 00986d48bb646daab659503ad3a713919865f32d |
|
26-Feb-2016 |
Derek Murray <mrry@google.com> |
Initial version of the open-source distributed TensorFlow runtime.
This includes a gRPC server (grpc_tensorflow_server) that can serve as both
the master of a distributed TensorFlow computation, and an individual worker
in the computation. The GrpcSession class is included to allow client programs
(including Python clients) to interact with a server.
See tensorflow/core/distributed_runtime/README.md for usage instructions.
This change partially addresses issue #23.
Change: 115634191
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| bce6216610d57f8f4b1e9e79836737df109c4e42 |
|
25-Feb-2016 |
Vincent Vanhoucke <vanhoucke@google.com> |
Switch nn.moments() to using a one-pass stable algorithm.
Helps with: https://github.com/tensorflow/tensorflow/issues/917
Also fixes https://github.com/tensorflow/tensorflow/issues/1162
The main benefit is that the computation of the sufficient statistics is now decoupled of the aggregation of the moments, which means that if you want to perform the accumulation incrementally, you don't have to keep all the inputs around, and can instead keep the much more compact sum and sum-of-squares. Accumulation could also be performed locally if you aggregate across multiple devices.
Computing sum and sum-of-squares can also theoretically be performed in parallel now.
Tested running inception: same performance, same step time.
Batch normalization benchmark is a bit faster on CPU, a bit slower on GPU:
Before:
cpu shape:4/3 #layers:10 mode:py scale:True train:False - 1.139310 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:False - 0.021970 secs
cpu shape:4/3 #layers:10 mode:py scale:True train:True - 2.767147 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:True - 0.074531 secs
cpu shape:4/3 #layers:10 mode:py scale:True train:False - 0.742835 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:False - 0.013473 secs
cpu shape:4/3 #layers:10 mode:py scale:True train:True - 1.738806 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:True - 0.052777 secs
cpu shape:2/1 #layers:10 mode:py scale:True train:False - 0.119180 secs
gpu shape:2/1 #layers:10 mode:py scale:True train:False - 0.011201 secs
cpu shape:2/1 #layers:10 mode:py scale:True train:True - 0.218297 secs
gpu shape:2/1 #layers:10 mode:py scale:True train:True - 0.048526 secs
After:
cpu shape:4/3 #layers:10 mode:py scale:True train:False - 0.998944 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:False - 0.025828 secs
cpu shape:4/3 #layers:10 mode:py scale:True train:True - 2.657428 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:True - 0.086614 secs
cpu shape:4/3 #layers:10 mode:py scale:True train:False - 0.603137 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:False - 0.017668 secs
cpu shape:4/3 #layers:10 mode:py scale:True train:True - 1.519533 secs
gpu shape:4/3 #layers:10 mode:py scale:True train:True - 0.055214 secs
cpu shape:2/1 #layers:10 mode:py scale:True train:False - 0.071344 secs
gpu shape:2/1 #layers:10 mode:py scale:True train:False - 0.016440 secs
cpu shape:2/1 #layers:10 mode:py scale:True train:True - 0.222093 secs
gpu shape:2/1 #layers:10 mode:py scale:True train:True - 0.039967 secs
Change: 115507032
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 4afef14f02682433f661b31ba57018d4ac6bdc18 |
|
24-Feb-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Support leaving the offset (beta) parameter out in batch_normalization, in which case no offset will be added after normalization.
Change: 115489328
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 369ca60ee6f025ecd29ab21268bb1e4aaeaa3995 |
|
19-Feb-2016 |
Vincent Vanhoucke <vanhoucke@google.com> |
Loosen test tolerances which seem to fail on Mac.
Change: 115027725
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 03869e6e9cfac85d3e0f87ded01f4dd8de10ed67 |
|
18-Feb-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Add utility function for creating partitioned variables.
Partitioned variables represent a single logical Variable across potentially multiple devices. They are saved and restored as if only a single Variable had been created, and the partitioning can change between save and restore.
Change: 114905595
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 7760ce56fc3ab4ab8cdc408e29d8ad8b539c417e |
|
11-Feb-2016 |
Josh Levenberg <josh11b@tensorflow.org> |
Get rid of some import cruft.
Change: 114374558
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| d53f06b6a3c77160ad955b718a04260674752298 |
|
07-Feb-2016 |
Vincent Vanhoucke <vanhoucke@google.com> |
https://github.com/tensorflow/tensorflow/issues/999
nn.moments()
- fix to handle axes that can't be automatically broadcast.
- add a keep_dims optional argument to returned means and variances with the same dimensionality as the inputs.
Change: 114056974
/external/tensorflow/tensorflow/python/ops/nn_test.py
|
| 9edd13a9c96edd9b7fdca71f694bb045f265ad66 |
|
06-Jan-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Rename deconv2d to conv2d_transpose and expose publicly
deconv2d is a misleading name for an operation which is exactly the transpose
of conv2d. conv2d_transpose is much better.
Fixes #256.
Change: 111529902
/external/tensorflow/tensorflow/python/ops/nn_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/ops/nn_test.py
|
| 854f49bd43588c062b046384f239f64a3d819702 |
|
25-Nov-2015 |
Manjunath Kudlur <keveman@gmail.com> |
TensorFlow: Upstream changes to git
Changes:
- Updates to docs
- Several changes for Python 3 compatibility
- Added license headers
Base CL: 108710566
/external/tensorflow/tensorflow/python/ops/nn_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/ops/nn_test.py
|
| d50565b35e886e7c3a201ea2f088790ed4b28de4 |
|
13-Nov-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: Upstream changes from afternoon.
Changes:
- Ptrdiff -> DenseIndex change by @jiayq
- Fix to scoping the logging in logging.py by @dga
- Improvement to Conv2DBackpropFilter on CPU by Andy
- Remove lookup table wrappers for the time being (wasn't in our
public API yet) by Yukata
- Add a check similar to numpy to make sure the user isn't in the
tensorflow src directory by @vrv
- More changes for python 3 compat by @girving
- Make dropout preserve shape info from input (@mrry)
- Significant speed improvements by @zheng-xq to BFC allocator to bring
on par (CPU overhead-wise) to the region allocator. Make BFC
allocator the default now that it's working well for a variety
of models.
- Fix a bunch of typos reported by users (@vrv)
- Enable concat for bfloat16 on GPU by Ashish.
Base CL: 107733123
/external/tensorflow/tensorflow/python/ops/nn_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/ops/nn_test.py
|
| 61d3a958d6d83cb6037490d933b47621cc4009cc |
|
09-Nov-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: Initial steps towards python3 support, some documentation
bug fixes -- reindents to 2 for some of the files to match our internal
requirements.
Thanks to Martin Andrews for the basic_usage.md suggested fix via
Gerrit.
Base CL: 107394029
/external/tensorflow/tensorflow/python/ops/nn_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/ops/nn_test.py
|