- 04 11月, 2016 36 次提交
-
-
由 A. Unique TensorFlower 提交于
Change: 138137511
-
由 A. Unique TensorFlower 提交于
or later no long has distribution.is_pure attribute. This change will cause the pip package tags to change based on the system it is built on. Update pip.sh to rename mac pip files(we already rename for linux) so pip names still matches our upload to gs script. Change: 138136631
-
由 A. Unique TensorFlower 提交于
Change: 138136264
-
由 A. Unique TensorFlower 提交于
Change: 138135876
-
由 Illia Polosukhin 提交于
- Estimator that supports input_fn only (x,y,batch_size are depreacated) - SKCompat wrapper that takes Estimator and has x,y,batch_size arguments. See updated tests for example usage. Change: 138134485
-
由 Andrew Harp 提交于
Change: 138133395
-
由 Patrick Nguyen 提交于
Change: 138131457
-
由 A. Unique TensorFlower 提交于
Change: 138131227
-
由 A. Unique TensorFlower 提交于
Change: 138130752
-
由 Illia Polosukhin 提交于
Change: 138128703
-
由 Xiaoqiang Zheng 提交于
added in a separate CL. For now, the fused implementation can be chosen with the environmental variable "TF_ADJUST_HUE_FUSED=1" Before the CL: benchmarkAdjustHue_299_299_3_cpu1 step_time: 15951.86 us benchmarkAdjustHue_299_299_3_cpu_all step_time: 2732.68 us After the CL: benchmarkAdjustHue_299_299_3_cpu1 step_time: 2450.50 us benchmarkAdjustHue_299_299_3_cpu_all step_time: 399.18 us Change: 138128641
-
由 Andrew Harp 提交于
Change: 138126747
-
由 A. Unique TensorFlower 提交于
Change: 138125802
-
由 A. Unique TensorFlower 提交于
Change: 138123658
-
由 A. Unique TensorFlower 提交于
Change: 138121817
-
由 A. Unique TensorFlower 提交于
Change: 138115819
-
由 Dan Mané 提交于
Change: 138108172
-
由 Dan Smilkov 提交于
Also: - Ignore checkpoints/configs that do not have any 2D tensors. - Shield main TensorBoard from plugin failures. Change: 138106394
-
由 Vijay Vasudevan 提交于
session.h/session_options.h and some other headers that are supposed to be publicly available. Fixes #5380 Change: 138104302
-
由 Andrew Harp 提交于
Change: 138103022
-
由 Jonathan Hseu 提交于
Change: 138101249
-
由 Sukriti Ramesh 提交于
Change: 138100290
-
由 A. Unique TensorFlower 提交于
Change: 138099098
-
由 Patrick Nguyen 提交于
Change: 138098284
-
由 Patrick Nguyen 提交于
Change: 138093178
-
由 Andrew Harp 提交于
Refactor Android demo to put activity specific code into ClassifierActivity to simplify adding new demos. Also add some utility classes and debug rendering (tap to enable). Change: 138091760
-
由 A. Unique TensorFlower 提交于
Change: 138090265
-
由 A. Unique TensorFlower 提交于
Change: 138090031
-
由 Andrew Harp 提交于
Change: 138089751
-
由 Vijay Vasudevan 提交于
Add shape function for FakeQuant gradient ops. Add a test for one of the more complicated ones. Change: 138087659
-
由 A. Unique TensorFlower 提交于
Change: 138083939
-
由 Benoit Steiner 提交于
Change: 138079985
-
由 Illia Polosukhin 提交于
* tf.all_variables to tf.global_variables * tf.VARIABLES -> tf.GLOBAL_VARIABLES * tf.initialize_all_variables -> tf.global_variable_initializers Change: 138078998
-
由 A. Unique TensorFlower 提交于
Change: 138078360
-
由 A. Unique TensorFlower 提交于
C++ Gradients: Adds gradient functions for some nn ops. Adds functionality to gradient checker to enable explicit initial values for some tensors. Change: 138078095
-
由 A. Unique TensorFlower 提交于
Benchmarks: Benchmark Time(ns) CPU(ns) Iterations -------------------------------------------------------------------------- BM_PTruncatedNormal_gpu_1000_1000 4632369 5175938 100 184.251M items/s BM_PTruncatedNormal_2SD_gpu_10000_100 2849437 3368804 206 283.090M items/s BM_PTruncatedNormal_OneTail_gpu_10000_100 3300317 3905713 179 244.174M items/s Change: 138074670
-
- 03 11月, 2016 4 次提交
-
-
由 A. Unique TensorFlower 提交于
a) EIGEN_USE_NONBLOCKING_THREAD_POOL no longer exist. The non-blocking threadpool is the default (It can be disabled by the symbol EIGEN_USE_SIMPLE_THREAD_POOL). b) An eigen_initialize.cc file is no longer needed due to recent fixes for thread safety in Eigen. Change: 138074328
-
由 A. Unique TensorFlower 提交于
Change: 138037407
-
由 Dan Mané 提交于
Right now, the names of all of the checked runs are stored in the url bar. This means that when TensorBoard first loads, it generates a (sometimes very long) url. With this change, TensorBoard stores the names of all the disabled runs. This has identical behavior except that the url bar will be short if all the runs are checked, and long if all the runs are unchecked. As a general invariant, I think loading a TensorBoard without touching the UI should always generate a short url (just /scalars maybe), and the url should grow long only if you add in settings. Change: 138032618
-
由 A. Unique TensorFlower 提交于
Change: 138030455
-