- 12 1月, 2017 3 次提交
-
-
由 Andrew Harp 提交于
Change: 144216903
-
由 Andrew Selle 提交于
tf.negative, tf.multiply, tf.subtract are the new names - Also enabled deprecation warning (to be completely removed by friday) Change: 144215355
-
由 Peter Hawkins 提交于
Change: 144211572
-
- 11 1月, 2017 36 次提交
-
-
由 Andrew Harp 提交于
Build and expose benchmark_model binary as part of Android nightly build, and also run build_all_android.sh for better build coverage. Change: 144158273
-
由 Andrew Harp 提交于
Change: 144158100
-
由 A. Unique TensorFlower 提交于
Change: 144155792
-
由 Andrew Harp 提交于
Android demo: Add support for YOLO-based object detection to object tracking demo. Set USE_YOLO boolean flag in DetectorActivity.java to true and add model according to directions there to enable. Change: 144155665
-
由 A. Unique TensorFlower 提交于
Change: 144153795
-
由 A. Unique TensorFlower 提交于
Change: 144152960
-
由 A. Unique TensorFlower 提交于
is active. Before the tfevents file was created even if the hook is never installed, which confuses Tensorboard and can cause it to watch the wrong file and not display updates. Change: 144152846
-
由 Lukasz Kaiser 提交于
Change: 144152550
-
由 A. Unique TensorFlower 提交于
Change: 144152544
-
由 Gunhan Gulsoy 提交于
Change: 144150966
-
由 Vijay Vasudevan 提交于
Change: 144145884
-
由 Pete Warden 提交于
Change: 144145086
-
由 A. Unique TensorFlower 提交于
Change: 144144091
-
由 Joshua V. Dillon 提交于
Change: 144144041
-
由 Mustafa Ispir 提交于
Change: 144142420
-
由 Andrew Selle 提交于
- Prevent tf.divide from using operator overloads which cannot pass names - Use dummy python class to properly delegate to the right division mode (i.e. python2 vs python3 vs from __future__ import divide) Fixes #6741 Change: 144140254
-
由 A. Unique TensorFlower 提交于
set tf.concat_v2 to be an alias to tf.concat and remove concat_v2 doc. Change: 144139939
-
由 Andrew Harp 提交于
Remove std:: from std::round() calls in benchmark_model.cc to avoid issue in Android ndk where std::round is not found: https://code.google.com/p/android/issues/detail?id=54418 Change: 144137666
-
由 A. Unique TensorFlower 提交于
Change: 144133445
-
由 A. Unique TensorFlower 提交于
op. Change: 144130931
-
由 Toby Boyd 提交于
Change: 144130687
-
由 Andrew Harp 提交于
Change: 144127608
-
由 Mustafa Ispir 提交于
Removed graph_action usage. Change: 144126485
-
由 Andrew Selle 提交于
This script currently is minimally tested. It is a work in progress currently. Change: 144125570
-
由 A. Unique TensorFlower 提交于
[XLA] Compute source indices in each common factor of reshape from target indices in same common factor Change: 144124189
-
由 David Majnemer 提交于
Change: 144124103
-
由 A. Unique TensorFlower 提交于
Change: 144117353
-
由 Sergio Guadarrama 提交于
Change: 144114955
-
由 Joshua V. Dillon 提交于
Change: 144112386
-
由 Yuan Yu 提交于
Set up the generated gradient while loops properly so that `None` is propagated when using tf.stop_gradient. Change: 144108792
-
由 Shanqing Cai 提交于
Change: 144108181
-
由 Mustafa Ispir 提交于
Change: 144101397
-
由 Asim Shankar 提交于
And fix a typo. Another step in the journey that is #5. Change: 144100664
-
由 Andrew Harp 提交于
Change: 144094196
-
由 Shanqing Cai 提交于
Change: 144094154
-
由 A. Unique TensorFlower 提交于
Change: 144086694
-
- 10 1月, 2017 1 次提交
-
-
由 Zongheng Yang 提交于
This hopefully addresses a potential race when training is not set up correctly: for instance, multiple training programs are running concurrently, but with the same log directory (used in tf.train.Saver()'s save_path). Change: 144078857
-