aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/eager
Commit message (Collapse)AuthorAge
...
* | Split checkpoint management utility functions out of saver.pyGravatar Allen Lavoie2018-08-02
| | | | | | | | | | | | Pure refactor, in preparation for adding a higher level checkpoint management utility. This utility will also need to work with the Checkpoint proto, and globbing it on to saver.py seems dirty. PiperOrigin-RevId: 207179646
| * Removed redundant tf.exp on predictions in evaluate methodGravatar Ruizhi2018-08-02
| | | | | | In the `evaluate` method, the usage of `tf.exp` seems redundant here, since `predictions` itself is the logits output of a dense layer in decoder. Therefore, I removed `tf.exp` so that `tf.multinomial` is applied on the logits directly, which agrees better with the proper usage of `tf.multinomial`.
* | [Checkpointable] Make Iterator checkpointable.Gravatar Shivani Agrawal2018-08-01
| | | | | | | | | | | | Use object-based save/restore to make dataset/iterator checkpointable in both graph as well as eager mode. PiperOrigin-RevId: 206998349
* | Update users of `tf.contrib.data.batch_and_drop_remainder(n)` to ↵Gravatar Derek Murray2018-08-01
| | | | | | | | | | | | `Dataset.batch(n, drop_remainder=True)`. PiperOrigin-RevId: 206953312
| * Fix shapes in comments of nmt_with_attention.ipynbGravatar Ruizhi2018-08-01
|/ | | It is a bit misleading and confusing that the output shape of decoder is currently commented as `(batch_size * max_length, vocab)`. However the correct shape should be `(batch_size * 1, vocab)`, since the input x of GRU layer has shape == `(batch_size, 1, embedding_dim + hidden_size)`.
* Remove logging.Gravatar Xuechen Li2018-07-31
| | | | PiperOrigin-RevId: 206859887
* Add explicit cast for range()Gravatar A. Unique TensorFlower2018-07-31
| | | | PiperOrigin-RevId: 206855526
* Update TPU estimator script to work with ImageNet.Gravatar Xuechen Li2018-07-31
| | | | PiperOrigin-RevId: 206825954
* Add regularization loss for densenet training.Gravatar Xuechen Li2018-07-31
| | | | | | Fixes #21237 PiperOrigin-RevId: 206791616
* Merge pull request #21212 from rakeshchada:masterGravatar TensorFlower Gardener2018-07-31
|\ | | | | | | PiperOrigin-RevId: 206784422
* | Add tpu train parameters in config.Gravatar Xuechen Li2018-07-30
| | | | | | | | PiperOrigin-RevId: 206658934
* | Update readme to add descriptions of training on TPUs.Gravatar Xuechen Li2018-07-30
| | | | | | | | PiperOrigin-RevId: 206658188
* | Minor rephrase of readme.Gravatar Xuechen Li2018-07-30
| | | | | | | | PiperOrigin-RevId: 206643365
* | Use `model_dir` as argument name to align with other examples.Gravatar Xuechen Li2018-07-30
| | | | | | | | PiperOrigin-RevId: 206643314
* | Remove excessive split and concat.Gravatar Xuechen Li2018-07-30
| | | | | | | | PiperOrigin-RevId: 206630163
| * Moved total loss calculation logicGravatar Rakesh Chada2018-07-30
| |
| * Fix the assignment of `batch_loss`Gravatar Rakesh Chada2018-07-29
| |
* | Add batch norm updates for estimator.Gravatar Xuechen Li2018-07-28
| | | | | | | | PiperOrigin-RevId: 206459313
| * Fix the epoch loss calculationGravatar Rakesh Chada2018-07-28
|/
* Update backward pass to save memory in graph mode.Gravatar Xuechen Li2018-07-27
| | | | PiperOrigin-RevId: 206352708
* Update backward pass to save memory in eager mode.Gravatar Xuechen Li2018-07-26
| | | | PiperOrigin-RevId: 206249965
* Make model totally defunable.Gravatar Xuechen Li2018-07-26
| | | | PiperOrigin-RevId: 206192038
* Allow setting server_def on tf.enable_eager_executionGravatar Akshay Modi2018-07-26
| | | | PiperOrigin-RevId: 206179345
* Add ImageNet TFRecords reading boilerplate.Gravatar Xuechen Li2018-07-25
| | | | PiperOrigin-RevId: 206081636
* Changing the dataset linkGravatar A. Unique TensorFlower2018-07-25
| | | | PiperOrigin-RevId: 206018486
* Adding defunGravatar A. Unique TensorFlower2018-07-24
| | | | PiperOrigin-RevId: 205901720
* Add dataset specific parameters in config file.Gravatar Xuechen Li2018-07-24
| | | | PiperOrigin-RevId: 205898175
* Use metric's dtype to represent matchesGravatar Igor Ganichev2018-07-24
| | | | | | | Before this change, even when the actual metric is using float32, we used float64 to represent boolean matches. PiperOrigin-RevId: 205891351
* Improvement resource variable documentation.Gravatar Alexandre Passos2018-07-24
| | | | | | | | Removes stale mentions of how to construct them from docstrings and adds a new section to the programmer's guide explaining resource handles and resource variable semantics. PiperOrigin-RevId: 205842012
* Add README for l2hmc.Gravatar Xuechen Li2018-07-23
| | | | PiperOrigin-RevId: 205733306
* Add defun option.Gravatar Xuechen Li2018-07-23
| | | | PiperOrigin-RevId: 205732970
* Further simplify backward pass.Gravatar Xuechen Li2018-07-23
| | | | PiperOrigin-RevId: 205728836
* Add main script training sampler on 2D energy landscapes.Gravatar Xuechen Li2018-07-23
| | | | PiperOrigin-RevId: 205707483
* Refactor imagenet configurations.Gravatar Xuechen Li2018-07-23
| | | | PiperOrigin-RevId: 205675401
* BEGIN_PUBLICGravatar Xuechen Li2018-07-20
| | | | | | | | Fix broken benchmark. END_PUBLIC Automated rollback of commit 8741006018326350467fe86785d98963ff9e983e PiperOrigin-RevId: 205480787
* Automated rollback of commit 265292420de30f24805d28886d403dc42d3685b3Gravatar A. Unique TensorFlower2018-07-20
| | | | PiperOrigin-RevId: 205472990
* Add estimator and TPU estimator training script.Gravatar Xuechen Li2018-07-20
| | | | PiperOrigin-RevId: 205446896
* fixing some nitsGravatar A. Unique TensorFlower2018-07-20
| | | | PiperOrigin-RevId: 205416917
* Fixing bug where in Eager mode datasets not intended to be on the GPU end up ↵Gravatar Rohan Jain2018-07-19
| | | | | | being there. PiperOrigin-RevId: 205331171
* Making the Eager iterator use the new copy_to_device.Gravatar Rohan Jain2018-07-19
| | | | | | This CL gets rid of the forced placement of all eager datasets / iterators on the CPU since now we can have some datasets on the GPU. PiperOrigin-RevId: 205264791
* Small text changeGravatar A. Unique TensorFlower2018-07-19
| | | | PiperOrigin-RevId: 205264291
* eager guide: s/tfe.Checkpoint/tf.train.Checkpoint/Gravatar Asim Shankar2018-07-19
| | | | PiperOrigin-RevId: 205248470
* DCGAN using tf.keras and eager execution.Gravatar A. Unique TensorFlower2018-07-18
| | | | PiperOrigin-RevId: 205149061
* Allows constructing resource variables from tf.Variable.Gravatar Alexandre Passos2018-07-18
| | | | | | | | Also adds arguments to control distributed aggregation to the tf.Variable constructor. Removes tfe.Variable from examples as it's now unnecessary. PiperOrigin-RevId: 205096552
* Benchmark for densenet model under eager and graph execution.Gravatar A. Unique TensorFlower2018-07-17
| | | | PiperOrigin-RevId: 204922359
* Replacing persistent tape with 2 different tapes for generator and ↵Gravatar A. Unique TensorFlower2018-07-12
| | | | | | discriminator. PiperOrigin-RevId: 204407120
* Add a test to eagerly run ops on multiple TPU coresGravatar Igor Ganichev2018-07-12
| | | | PiperOrigin-RevId: 204354687
* changed the image link.Gravatar A. Unique TensorFlower2018-07-12
| | | | PiperOrigin-RevId: 204330664
* Fix clickable buttons on notebooksGravatar Billy Lamberta2018-07-11
| | | | PiperOrigin-RevId: 204230306
* Merge pull request #20429 from yashk2810:image_captioningGravatar TensorFlower Gardener2018-07-10
|\ | | | | | | PiperOrigin-RevId: 204018286
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-25 22:49:59 +0000