- 01 8月, 2016 1 次提交
-
-
由 A. Unique TensorFlower 提交于
Change: 128958134
-
- 26 7月, 2016 1 次提交
-
-
由 Martin Wicke 提交于
Change: 128401884
-
- 13 7月, 2016 3 次提交
-
-
由 Benoit Steiner 提交于
improvements for fp16 Added SpecialFunctions to the list of eigen headers TensorFlow depends on Change: 127264575
-
由 Vijay Vasudevan 提交于
Change: 127253427
-
由 Benoit Steiner 提交于
improvements for fp16 Change: 127233960
-
- 01 7月, 2016 1 次提交
-
-
由 A. Unique TensorFlower 提交于
Change: 126335170
-
- 28 6月, 2016 1 次提交
-
-
由 A. Unique TensorFlower 提交于
Adds a "currentThreadIndex" method to Eigen's ThreadPoolDevice. Use it to handle per-thread buffer allocation for the tileable executor without resorting to thread_local that is not fully supported on Android. Change: 126009029
-
- 24 6月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
Upgraded Eigen to the latest version that provides new scan operations. This will enable the implementation of the cumsum operation in TensorFlow Change: 125697517
-
- 09 6月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
performance of the toy mnist training by 1 order of magnitude Change: 124374286
-
- 08 6月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
NEW BM_fullReduction/10 4591 4595 153149 20.8M items/s BM_fullReduction/64 5073 5075 100000 770.0M items/s BM_fullReduction/512 9067 9070 75263 26.9G items/s BM_fullReduction/4k 243984 244125 2868 64.0G items/s BM_fullReduction/5k 359125 359273 1951 64.8G items/s OLD BM_fullReduction/10 9085 9087 74395 10.5M items/s BM_fullReduction/64 9478 9478 72014 412.1M items/s BM_fullReduction/512 14643 14646 46902 16.7G items/s BM_fullReduction/4k 260338 260384 2678 60.0G items/s BM_fullReduction/5k 385076 385178 1818 60.5G items/s Change: 124290852
-
- 07 6月, 2016 2 次提交
-
-
由 Benoit Steiner 提交于
gradients, some variants etc.). Change: 124197406
-
由 Martin Wicke 提交于
Change: 124183870
-
- 03 6月, 2016 3 次提交
-
-
由 A. Unique TensorFlower 提交于
gradients, some variants etc.). Change: 123967787
-
由 Benoit Steiner 提交于
gradients, some variants etc.). Change: 123967117
-
由 A. Unique TensorFlower 提交于
Change: 123901292
-
- 01 6月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
Change: 123659102
-
- 28 5月, 2016 2 次提交
-
-
由 Kiril Gorovoy 提交于
Change: 123468144
-
由 A. Unique TensorFlower 提交于
tf.train.Example. Change: 123445810
-
- 26 5月, 2016 2 次提交
-
-
由 A. Unique TensorFlower 提交于
up to the OSS build yet, we're working on it.) Change: 123248081
-
由 Benoit Steiner 提交于
Change: 123238579
-
- 24 5月, 2016 1 次提交
-
-
由 Derek Murray 提交于
Change: 123026122
-
- 20 5月, 2016 1 次提交
-
-
由 Alexey Surkov 提交于
Implements an authentication mechanism based on Application Default Credentials: https://developers.google.com/identity/protocols/application-default-credentials https://developers.google.com/identity/protocols/OAuth2ServiceAccount Change: 122741738
-
- 17 5月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
with many cpu cores For example, the wall time for the following tutorial went down from 13m35 to 5m27: bazel run -c opt --copt=-mavx tensorflow/examples/tutorials/word2vec/word2vec_basic Change: 122462177
-
- 14 5月, 2016 3 次提交
-
-
由 A. Unique TensorFlower 提交于
and remove the cuda_crosstool_condition build condition. Now if_cuda is just using_nvcc || using_gcudacc. Change: 122291892
-
由 A. Unique TensorFlower 提交于
This is unused at the moment, but will eventually let us build CUDA code with vanilla clang. Change: 122289910
-
由 A. Unique TensorFlower 提交于
This has no practical effect, as CUDA builds are always with nvcc, but it lets us modify the build config rule //third_party/gpus/cuda:using_nvcc so it returns true, rather than false, for CUDA builds. Change: 122288952
-
- 13 5月, 2016 2 次提交
-
-
由 Benoit Steiner 提交于
Change: 122192081
-
由 Benoit Steiner 提交于
by about 3 orders of magnitude as well as some partial reductions by 30% when using cuda 7.5 or above Change: 122191448
-
- 06 5月, 2016 1 次提交
-
-
由 A. Unique TensorFlower 提交于
Change: 121586635
-
- 28 4月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
gpus Updated the check numerics code to make it compatible with fp16 Change: 120980302
-
- 26 4月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
Change: 120739269
-
- 22 4月, 2016 2 次提交
-
-
由 A. Unique TensorFlower 提交于
tensorflow: switch to eigen thread pool This is first step of switching tensorflow to the new non-blocking thread pool in eigen. Change: 120510292
-
由 Benoit Steiner 提交于
Change: 120505517
-
- 21 4月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
offered by AWS Change: 120369420
-
- 15 4月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
sigmoid of fp16 and introduces a condition estimator. Change: 119907721
-
- 14 4月, 2016 3 次提交
-
-
由 Benoit Steiner 提交于
Change: 119850987
-
由 Benoit Steiner 提交于
improvements for fp16 Change: 119771118
-
由 A. Unique TensorFlower 提交于
Change: 119768540
-
- 10 4月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
Change: 119458778
-
- 09 4月, 2016 1 次提交
-
-
由 Benoit Steiner 提交于
as well as fp16 Change: 119398881
-