aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/layers
Commit message (Collapse)AuthorAge
* Automated rollback of commit 425e96f3ae4eb338268e3738260f9d79e4bdd893. ↵Gravatar A. Unique TensorFlower2018-09-27
| | | | | | Revert #20539. PiperOrigin-RevId: 214849875
* Updating the V2 variables API.Gravatar Alexandre Passos2018-09-27
| | | | PiperOrigin-RevId: 214824023
* Merge pull request #22350 from yongtang:22295-gradient_multipliersGravatar TensorFlower Gardener2018-09-24
|\ | | | | | | PiperOrigin-RevId: 214297796
* \ Merge pull request #18567 from imsheridan:fix_expand_dimsGravatar TensorFlower Gardener2018-09-24
|\ \ | | | | | | | | | PiperOrigin-RevId: 214278672
* \ \ Merge pull request #20539 from candyzone:masterGravatar TensorFlower Gardener2018-09-22
|\ \ \ | | | | | | | | | | | | PiperOrigin-RevId: 214121495
| | * \ Merge branch 'master' into fix_expand_dimsGravatar Martin Wicke2018-09-22
| | |\ \ | |_|/ / |/| | |
| * | | Fix lint errorsGravatar Allen Lavoie2018-09-21
| | | |
| | | * Fix pylint issueGravatar Yong Tang2018-09-21
| | | | | | | | | | | | | | | | Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* | | | Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 213326581
| | | * Add additional test casesGravatar Yong Tang2018-09-17
| | | | | | | | | | | | | | | | Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
| | | * Allow different dtype of TensorGravatar Yong Tang2018-09-17
| | | | | | | | | | | | | | | | Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
| | | * Add test cases to allow gradient_multipliers passed as tensorGravatar Yong Tang2018-09-17
| | | | | | | | | | | | | | | | Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
| | | * Support gradient_multipliers as tensor for optimize_lossGravatar Yong Tang2018-09-17
| |_|/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fix tries to address the issue raised in 22295 where gradient_multipliers for tf.contrib.layers.optimize_loss() does not support tensor as input. This fix update the optimize_loss to allow gradient_multipliers passed as dict of tensors. This fix fixes 22295. Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* | | Merge pull request #22031 from rogerxcn:masterGravatar TensorFlower Gardener2018-09-10
|\ \ \ | | | | | | | | | | | | PiperOrigin-RevId: 212399243
* | | | Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 212336321
* | | | Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 212336206
* | | | Fix flakiness in ConvolutionInPlaneTest.testVertConvWithBlankImage byGravatar A. Unique TensorFlower2018-09-04
| | | | | | | | | | | | | | | | | | | | | | | | switching from assertAllEqual to assertAllClose. PiperOrigin-RevId: 211543406
| * | | Fix issues in maxout layerGravatar Roger Xin2018-09-03
|/ / /
* | | Usability improvements to @recompute_gradGravatar A. Unique TensorFlower2018-08-31
| | | | | | | | | | | | | | | | | | | | | Error if fn closes over Tensor or Variable (not always detectable) Allow None gradients to some inputs (filter out Nones before control_deps) PiperOrigin-RevId: 211162615
* | | Updates the warning of default initializer in deprecated embedding_column.Gravatar Mustafa Ispir2018-08-31
| | | | | | | | | | | | PiperOrigin-RevId: 211145700
* | | Move from deprecated self.test_session() to self.session() when a graph is set.Gravatar A. Unique TensorFlower2018-08-21
| | | | | | | | | | | | | | | | | | self.test_session() has been deprecated in cl/208545396 as its behavior confuses readers of the test. Moving to self.session() instead. PiperOrigin-RevId: 209696110
* | | Add one pass algorithm option to calculate the mean and variance in ↵Gravatar A. Unique TensorFlower2018-08-20
| | | | | | | | | | | | | | | | | | group_norm. Fix normalization test in test fusion. PiperOrigin-RevId: 209534762
* | | Ensure that functools.wraps is not called on functools.partial objects in ↵Gravatar A. Unique TensorFlower2018-08-20
| | | | | | | | | | | | | | | | | | rev_block. PiperOrigin-RevId: 209524010
* | | Fix the issue of losing last dimension for one hot column with weightsGravatar A. Unique TensorFlower2018-08-20
| | | | | | | | | | | | PiperOrigin-RevId: 209485789
* | | Automated rollback of commit a1606d5e0f667fddd7f3f5705bda3aee5b3c2554Gravatar Gunhan Gulsoy2018-08-17
| | | | | | | | | | | | PiperOrigin-RevId: 209187557
* | | Add one pass algorithm option to calculate the mean and variance in group_normGravatar A. Unique TensorFlower2018-08-16
| | | | | | | | | | | | PiperOrigin-RevId: 209096783
* | | Remove magic-doc-links from code.Gravatar Mark Daoust2018-08-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change contains no code changes. Only doc-strings. We can't use relative links in code files, so we don't have much choice but to link to tensorflow.org/ The deleted links were to docs that no longer exist. PiperOrigin-RevId: 209019572
* | | Merge pull request #18955 from brettkoonce:minor_sp_04_28Gravatar TensorFlower Gardener2018-08-14
|\ \ \ | | | | | | | | | | | | PiperOrigin-RevId: 208733608
* | | | Fix flaky layers test.Gravatar Akshay Agrawal2018-08-10
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 208269820
| * | | contrib: minor spelling tweaksGravatar Brett Koonce2018-08-09
|/ / /
* | | Remove usage of magic-api-link syntax from source files.Gravatar Mark Daoust2018-08-09
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Back-ticks are now converted to links in the api_docs generator. With the new docs repo we're moving to simplify the docs pipeline, and make everything more readable. By doing this we no longer get test failures for symbols that don't exist (`tf.does_not_exist` will not get a link). There is also no way, not to set custom link text. That's okay. This is the result of the following regex replacement (+ a couple of manual edits.): re: @\{([^$].*?)(\$.+?)?} sub: `\1` Which does the following replacements: "@{tf.symbol}" --> "`tf.symbol`" "@{tf.symbol$link_text}" --> "`tf.symbol`" PiperOrigin-RevId: 208042358
* | | Set depth_multiplier default value, so that one can use separable_conv2d and ↵Gravatar A. Unique TensorFlower2018-08-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | conv2d interchangeably (conv2d does not have depth_multiplier arg). For example: if use_separable: conv_op = separable_conv2d else: conv_op = conv2d net = conv_op(....) Most common usage of separable conv use depth_multiplier=1, e.g. in MobileNet and Xception. PiperOrigin-RevId: 207741004
* | | Merge pull request #20554 from yongtang:20551-contrib.layers.conv1dGravatar TensorFlower Gardener2018-07-31
|\ \ \ | | | | | | | | | | | | PiperOrigin-RevId: 206825546
| | * | fix: No need to convert to tensor when using ResourceVariable in ↵Gravatar candy.dc2018-07-26
| |/ / |/| | | | | | | | | | | | | | embedding_lookup, because ResourceVariable support ResourceGather OP.
* | | Add synchronization and aggregation arguments to variable creation methods ↵Gravatar Yuefeng Zhou2018-07-22
| | | | | | | | | | | | | | | | | | in contrib/layers. PiperOrigin-RevId: 205588849
* | | Update recompute_grad to work for functions without variables.Gravatar A. Unique TensorFlower2018-07-09
| | | | | | | | | | | | | | | | | | Wrap recompute inputs in identity to avoid exponential graph traversal in gradients. PiperOrigin-RevId: 203776272
| * | Expose conv1d in allowed symbol.Gravatar Yong Tang2018-07-05
| | | | | | | | | | | | Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
| * | Alias convolution1d with conv1d in tf.contrib.layersGravatar Yong Tang2018-07-05
| | | | | | | | | | | | Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
| * | Expose `tf.contrib.layers.conv1d` like conv2d/conv3dGravatar Yong Tang2018-07-05
|/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fix tries to address the issue raised in 20551 where `tf.contrib.layers.conv1d` is not exposed. In tf.contrib.layers even though conv1d/conv2d/conv3d are all available, the conv1d was not exposed like conv2d/conv3d. This fix exposed `tf.contrib.layers.conv1d` so that it is consistent with `tf.contrib.layers.conv2d` and `tf.contrib.layers.conv3d`. This fix fixes 20551. Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* | Merge pull request #18771 from facaiy/CLN/clean_duplicate_safe_embedding_fnGravatar Martin Wicke2018-07-02
|\ \ | | | | | | make safe_embedding_lookup_sparse method public and clean duplicate codes
* \ \ Merge commit for internal changesGravatar Mingxing Tan2018-06-23
|\ \ \
| * | | For separable_conv2d, add an optional pointwise_initializer parameter so ↵Gravatar A. Unique TensorFlower2018-06-22
| | | | | | | | | | | | | | | | | | | | | | | | that the pointwise weights can be initialized differently from the depthwise weights. PiperOrigin-RevId: 201705897
* | | | Merge commit for internal changesGravatar Akshay Modi2018-06-15
|\| | |
| * | | Documentation style fix.Gravatar Stanley Bileschi2018-06-13
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 200477609
* | | | Merge commit for internal changesGravatar Patrick Nguyen2018-06-07
|\| | |
| * | | Add convolution and convolution1d to the public APIGravatar A. Unique TensorFlower2018-06-07
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 199642103
* | | | Merge commit for internal changesGravatar Yifei Feng2018-05-29
|\| | |
| * | | Merge changes from github.Gravatar Yifei Feng2018-05-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Revert #18413. Too many internal test failures due to the name scope change caused by this change. Revert #18192. Cannot use re2::StringPiece internally. Need alternative for set call. Will pull and clean this up in a separate change. PiperOrigin-RevId: 197991247
| * | | Small fix so that GDN can run on TPUGravatar A. Unique TensorFlower2018-05-24
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 197959536
| * | | Add support for is_recompute optional kwarg to functions decorated with ↵Gravatar A. Unique TensorFlower2018-05-23
| | | | | | | | | | | | | | | | | | | | | | | | recompute_grad PiperOrigin-RevId: 197834316