- 27 7月, 2017 3 次提交
-
-
由 Amit Patankar 提交于
Update version to rc1.
-
由 Amit Patankar 提交于
-
由 Amit Patankar 提交于
-
- 26 7月, 2017 21 次提交
-
-
由 Amit Patankar 提交于
Cherrypicks
-
由 Amit Patankar 提交于
-
由 Amit Patankar 提交于
-
由 Olivia Nordquist 提交于
PiperOrigin-RevId: 162379684
-
由 Amit Patankar 提交于
-
由 Shanqing Cai 提交于
Format 1.3.0 release notes; added a new feature to tfdbg section
-
由 Jonathan Hseu 提交于
-
由 Eugene Brevdo 提交于
Fixes #11666. PiperOrigin-RevId: 162776187
-
由 A. Unique TensorFlower 提交于
is specified but output_buffer_size is not: ValueError: Tensor conversion requested dtype int64 for Tensor with dtype int32: 'Tensor("num_threads:0", shape=(), dtype=int32)' PiperOrigin-RevId: 162766317
-
由 Benoit Steiner 提交于
preserving them when they're on the sender side. PiperOrigin-RevId: 162698209
-
由 Brennan Saeta 提交于
The base_rendezvous_mgr handles transferring a tensor using DMAs to non-"host" devices such as GPUs in the SameWorkerRecvDone function. This function would use the worker_env's device_mgr to obtain a pointer to the relevant Device (using the LookupDevice call). In the ClusterSpec-propagation world, using the environment's device_set means that the devices are not renamed, often resulting in devices not being found. This change updates BaseRemoteRendezvous to use the WorkerSession stored when the BaseRemoteRendezvous is initialized. The WorkerSession has a pointer to a DeviceMgr that contains the appropriately renamed devices for the given session the Rendezvous is associated with. Note: because we have a fast-path host-device-only copy, the original bug does not show up when using 2 CPU devices. I have added a test to ensure that transferring between 2 CPU devices works in a ClusterSpec propagation session, but note that this test does not actually reproduce the motivating bug. In the process of writing a test for the original bug, I discovered another latent bug in ClusterSpec propagation where if there were 2 CPU devices (i.e. due to explicit server configuration to have 2 CPU devices), a DCHECK could be triggered. The Master::CreateSession would call `device_set->set_client_device` multiple times (once for each CPU device). PiperOrigin-RevId: 162680163
-
由 Dandelion Man? 提交于
The display_name will be used to display the series in TensorBoard, in lieu of the tag, assuming that it is specified. (When it is not specified, behavior will stay the same.) The summary_description will allow the user to write a longform readable description of the summary series for display in TensorBoard. Markdown will be supported. This will make it possible for TensorBoard-2 summary ops to give the user direct control over the display name, solving https://github.com/tensorflow/tensorboard/issues/59. PiperOrigin-RevId: 162566261
-
由 Adam Roberts 提交于
Add multi-head attention capabilities to AttentionWrapper via the specification of multiple AttentionMechanisms. PiperOrigin-RevId: 162557562
-
由 RJ Ryan 提交于
PiperOrigin-RevId: 162523648
-
由 RJ Ryan 提交于
These are implemented using GPU-capable ops and have gradient support. PiperOrigin-RevId: 162511440
-
由 Mark Daoust 提交于
PiperOrigin-RevId: 162418535
-
由 A. Unique TensorFlower 提交于
PiperOrigin-RevId: 162406054
-
由 Jonathan Hseu 提交于
PiperOrigin-RevId: 162392717
-
由 Eugene Brevdo 提交于
Related issues: #11077, #11540. PiperOrigin-RevId: 162362081
-
由 Shanqing Cai 提交于
-
由 Amit Patankar 提交于
Cherry-pick GradientsDebugger from master to r1.3, with fixes to GPU build issues
-
- 25 7月, 2017 3 次提交
-
-
由 Shanqing Cai 提交于
Replace it with a new hidden op: array_ops.DebugGradientIdentity PiperOrigin-RevId: 162983526
-
由 Shanqing Cai 提交于
Also updates programmer_guide/debugger.md RELNOTES: TensorFlow Debugger (tfdbg) CLI: added ability to display feed_dict values to Session.run through the "pf" / "print_feed" command and clickable links in the curses UI. PiperOrigin-RevId: 162648456
-
由 Shanqing Cai 提交于
to allow retrieval of gradient tensors created by TensorFlow's automatic differentiation algorithm (i.e., tf.gradients and optimizer code that uses it). PiperOrigin-RevId: 161805516
-
- 23 7月, 2017 3 次提交
-
-
由 Amit Patankar 提交于
Add a note about tf.contrib.signal to the 1.3 release notes.
-
由 Amit Patankar 提交于
Add new feature description for tfdbg to r1.3 release notes
-
由 RJ Ryan 提交于
-
- 22 7月, 2017 1 次提交
-
-
由 Shanqing Cai 提交于
-
- 21 7月, 2017 2 次提交
-
-
由 Amit Patankar 提交于
Updating RELEASE notes.
-
由 Amit Patankar 提交于
-
- 20 7月, 2017 5 次提交
-
-
由 Amit Patankar 提交于
Updating the RELNOTES for 1.3.0rc0.
-
由 Amit Patankar 提交于
-
由 Amit Patankar 提交于
-
由 Amit Patankar 提交于
-
由 Amit Patankar 提交于
-
- 19 7月, 2017 2 次提交
-
-
由 Amit Patankar 提交于
Revert commits 1.3.0rc0
-
由 Amit Patankar 提交于
This reverts commit 31f1375c.
-