- 07 6月, 2018 4 次提交
-
-
由 Francois Chollet 提交于
-
由 Bohumír Zámečník 提交于
* Add a unit test for CuDNNGRU conversion with TimeDistributed. * Extract duplicated function convert_model() to _convert_model_weights(). * #10356 Convert weights of CuDNN/plain RNN nested in TimeDistributed. Same case as for Bidirectional, except that in TimeDistributed there's only one nested layer instead of two.
-
由 Taehoon Lee 提交于
-
由 Francois Chollet 提交于
-
- 06 6月, 2018 5 次提交
-
-
由 Taehoon Lee 提交于
-
由 Francois Chollet 提交于
-
由 Francois Chollet 提交于
-
由 tiferet 提交于
-
由 KuzMenachem 提交于
* Fix issue when working in conjunction with TimeSeriesGenerator The TimeSeriesGenerator class uses xrange through six which caused an IndexError * Add test * Add corresponding test * Fix for python3 * Simplified code * Fix indent * Fix test
-
- 05 6月, 2018 3 次提交
-
-
由 David Schwertfeger 提交于
* Embed layer-outputs rather than layer-weights in TensorBoard callback * Update docstring and allow multiple inputs * Fix tests * Renaming * Set learning phase * Compute embeddings in batches * Pass embedding data explicitly * Actually process embeddings in batches * Allow multiple inputs and validate input data * Add example * Delete utils.py * Revert uncorrectly resolved merge-conflict * Minor renaming * Add comment clarifying the design choice
-
由 Tommi Koivisto 提交于
The max_value argument can not be used in a layer, except custom layer or Lambda. Hence, similarly to LeakyReLU or for example Softmax, this PR adds a layer for ReLU, enabling also a capped ReLU to be used.
-
由 Taehoon Lee 提交于
* Reduce tests for applications * Make selection over all models random
-
- 03 6月, 2018 2 次提交
-
-
由 Anirudh Swaminathan 提交于
* Add depth as third dimension in docstring of Conv3DTranspose in convolutional.py in keras.layers
-
由 Taehoon Lee 提交于
-
- 02 6月, 2018 2 次提交
-
-
由 François Chollet 提交于
* Split `applications` and `preprocessing` modules. * Fix dependencies.
-
由 Zhengping Che 提交于
* equip TimeDistributed with mask and unspecified input length * fix bugs in theano. add test on timedistributed + masking * skip tests on cntk with multiple unspecified time lengths. * move static shape inference to theano_backend, add docstring, etc. * fix format
-
- 01 6月, 2018 1 次提交
-
-
由 fuzzythecat 提交于
Fixed doc
-
- 30 5月, 2018 3 次提交
-
-
由 fuzzythecat 提交于
* Add pooling option * Add pooling test
-
由 Taehoon Lee 提交于
-
由 Francois Chollet 提交于
-
- 29 5月, 2018 1 次提交
-
-
由 fuzzythecat 提交于
* Fix naming convention * Add missing doc
-
- 28 5月, 2018 1 次提交
-
-
由 r-kellerm 提交于
-
- 26 5月, 2018 3 次提交
-
-
由 Botty Dimanov 提交于
#10219
-
由 Wang, Zhiming 提交于
Non training Batch Norm operator has bad performance for it running into tensorflow's non fused batch norm API (#10207) * When use tensorflow as backend, let batch norm run into fused batch norm as much as possible, which has better performance. fix issue: http://github.com/keras-team/keras/issues/10058 * In Tensorflow backend, let batch norm call to FusedBatchNorm only NHWC format, also gamma and beta are not None. Test result: test env: with Tensorflow(commit a543d9471047ca3f6881c87105fcbe2cdff9207d Date: Thu May 10 17:43:30 2018, local build), python3.4, centos7.4 test cases: "pytest ./tests/keras/layers/normalization_test.py" <all passed> "pytest ./tests" <keep same result as without this commit's modification on BN> * fix code sytle. * 1. Add axis parameter in backend's batch_normalization functions. 2. Refine the batch_normalization function in tensorflow backend, Let's it call to fused batch norm as much as possible. Thanks the coments from fchollet. * Trigger * 1. add default value -1 for parameter axis in batch_normalization function in backend. 2. fix some code style. Thanks the comments from fchollet.
-
由 Stanley Bileschi 提交于
* Alphabetizes and adds to layers doc. * Responding to @cais comments * fix spacing. Remove in(out)bound_nodes
-
- 25 5月, 2018 3 次提交
-
-
由 Amir Alavi 提交于
Fix for #10265
-
由 jnphilipp 提交于
-
由 Stefano Cappellini 提交于
* Fixes automatic doc generation problem with indented lists. Adds a new test * Some style fixes on doc automatic generation files * Fixes a bad space in convolutional_recurrent.py * Changes the test_doc_auto_generation in order to include a doc string taken from the codebase. Allows text lines following nested lists
-
- 24 5月, 2018 3 次提交
-
-
由 Clemens Tolboom 提交于
Doc expects a list containing 2 tensors.
-
由 askskro 提交于
Softplus docstring missing a parenthesis.
-
由 Brian Nemsick 提交于
-
- 23 5月, 2018 2 次提交
-
-
由 vkk800 提交于
* Create get_random_transform and refactor * Fix style and add tests * Add more tests * Fix documentation error * Fix documentation style issue * add apply_affine_transform * document transformation dictionary * Doc style fix
-
由 Taehoon Lee 提交于
-
- 22 5月, 2018 2 次提交
-
-
由 brge17 提交于
* Remove Sequential.model deprecation warning * Remove dead line of code
-
由 Taehoon Lee 提交于
-
- 19 5月, 2018 1 次提交
-
-
由 ghostplant 提交于
New Benchmark by in-place split: >> keras.application.Resnet50 224x224x3 (NCWH; NVidia Tesla P100 x 4) input_shape = 3x224x224, batch_size = 96 x 4: 392(images/sec) => 417(images/sec) input_shape = 3x299x299, batch_size = 64 x 4: 229(images/sec) => 244(images/sec) input_shape = 3x224x224, batch_size = 8 x 4: 148(images/sec) => 163(images/sec) >> keras.application.InceptionV3 (NCWH; NVidia Tesla P100 x 4) input_shape = 3x224x224, batch_size = 128 x 4: 488(images/sec) => 526(images/sec) input_shape = 3x299x299, batch_size = 96 x 4: 270(images/sec) => 294(images/sec) input_shape = 3x224x224, batch_size = 8 x 4: 146(images/sec) => 158(images/sec) Signed-off-by: NCUI Wei <ghostplant@qq.com>
-
- 18 5月, 2018 4 次提交
-
-
由 Taehoon Lee 提交于
* Fix NASNet * Update weight files
-
由 ghostplant 提交于
Signed-off-by: NCUI Wei <ghostplant@qq.com>
-
由 Paige Bailey 提交于
-
由 Leonardo Taccari 提交于
-