- 25 Nov, 2019 2 commits
-
-
Sai Ganesh Bandiatmakuri authored
This will help general debugging by enabling custom execution with --benchmark_method_steps. E.g --benchmark_method_steps=train_steps=7 will run the benchmark for only 7 steps without modifying benchmark code. PiperOrigin-RevId: 282396875
-
Chen Chen authored
PiperOrigin-RevId: 282384787
-
- 22 Nov, 2019 2 commits
-
-
David Chen authored
PiperOrigin-RevId: 281977686
-
David Chen authored
PiperOrigin-RevId: 281908068
-
- 20 Nov, 2019 1 commit
-
-
Chen Chen authored
PiperOrigin-RevId: 281473612
-
- 12 Nov, 2019 1 commit
-
-
Rajagopal Ananthanarayanan authored
PiperOrigin-RevId: 280019807
-
- 11 Nov, 2019 2 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 279768972
-
Hongkun Yu authored
- Update classifier example. - Add new converted checkpoints. - Update benchmark, PiperOrigin-RevId: 279762797
-
- 08 Nov, 2019 1 commit
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 279361336
-
- 06 Nov, 2019 2 commits
-
-
Ayush Dubey authored
PiperOrigin-RevId: 278902210
-
Yeqing Li authored
PiperOrigin-RevId: 278756245
-
- 02 Nov, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 278028560
-
- 01 Nov, 2019 2 commits
-
-
Reed Wanderman-Milne authored
To test, I ran the following command: python resnet_ctl_imagenet_main.py --batch_size=2048 --data_dir ~/imagenet --datasets_num_private_threads=14 --epochs_between_evals=10 --model_dir ~/tmp_model_dir --clean --num_gpus=8 --train_epochs=90 --dtype=fp16 I got 76.15% final evaluation accuracy. PiperOrigin-RevId: 278010061
-
Hongkun Yu authored
PiperOrigin-RevId: 277992916
-
- 29 Oct, 2019 1 commit
-
-
David Chen authored
PiperOrigin-RevId: 277332242
-
- 28 Oct, 2019 1 commit
-
-
Zongwei Zhou authored
PiperOrigin-RevId: 277082247
-
- 22 Oct, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 275986990
-
- 18 Oct, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 275417626
-
- 17 Oct, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 275288636
-
- 16 Oct, 2019 3 commits
-
-
David Chen authored
PiperOrigin-RevId: 275103426
-
Yeqing Li authored
PiperOrigin-RevId: 275080469
-
Yeqing Li authored
PiperOrigin-RevId: 274921478
-
- 11 Oct, 2019 1 commit
-
-
A. Unique TensorFlower authored
Change summary directory and model checkpoint directory so that training via Keras Compile/Fit() and custom training loop is consistent. PiperOrigin-RevId: 274202793
-
- 10 Oct, 2019 1 commit
-
-
Yeqing Li authored
PiperOrigin-RevId: 274035928
-
- 09 Oct, 2019 1 commit
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 273795511
-
- 01 Oct, 2019 1 commit
-
-
David Chen authored
PiperOrigin-RevId: 272121528
-
- 26 Sep, 2019 1 commit
-
-
David Chen authored
PiperOrigin-RevId: 271422100
-
- 24 Sep, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 270817869
-
- 19 Sep, 2019 2 commits
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 270122397
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 270079001
-
- 18 Sep, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 269835917
-
- 13 Sep, 2019 1 commit
-
-
Ayush Dubey authored
PiperOrigin-RevId: 268836696
-
- 12 Sep, 2019 1 commit
-
-
Ayush Dubey authored
PiperOrigin-RevId: 268703439
-
- 29 Aug, 2019 1 commit
-
-
Reed Wanderman-Milne authored
The old infer_float32_policies policy will be removed from TensorFlow soon. To test convergence, I ran the Resnet50KerasAccuracy.benchmark_8_gpu_fp16 benchmark. I got an accuracy of 0.76037 and an exp_per_second of 6908. PiperOrigin-RevId: 266191126
-
- 28 Aug, 2019 1 commit
-
-
Adrian Kuegel authored
PiperOrigin-RevId: 265940771
-
- 22 Aug, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 264895439
-
- 21 Aug, 2019 1 commit
-
-
Ayush Dubey authored
PiperOrigin-RevId: 264511778
-
- 20 Aug, 2019 2 commits
-
-
Vinh Nguyen authored
-
Vinh Nguyen authored
using existing flag --fp16_implementation in official/utils/flags/_performance.py to enable automatic mixed precision
-
- 19 Aug, 2019 1 commit
-
-
Haoyu Zhang authored
Moving Keras ResNet models to `official/vision/image_classification` and benchmarks to `official/benchmark`. PiperOrigin-RevId: 264268533
-