Commits · 51fc02ae735c6297906814090340034bb1d574da · ModelZoo / ResNet50_tensorflow

"...internlm2-math-7b.git" did not exist on "938324473e471e760a28d37c177c9a8771353a41"

20 Dec, 2018 4 commits
- Add skip_eval flag and change to new optimizer · 51fc02ae
  Shining Sun authored Dec 20, 2018
  
  51fc02ae
- bug fixes and clean ups · 6f881f77
  Shining Sun authored Dec 20, 2018
  
  6f881f77
- Inlude the distribution_utils file · b1b4c805
  Shining Sun authored Dec 19, 2018
  
  b1b4c805
- Before all the data related change · 424c2045
  Shining Sun authored Dec 19, 2018
  
  424c2045
19 Dec, 2018 2 commits
- Commit before pull · 26a2c3b9
  Shining Sun authored Dec 19, 2018
  
  26a2c3b9
- Clean up keras_resnet_model and rename it to resnet56 · 42420ce8
  Shining Sun authored Dec 19, 2018
  
  42420ce8
18 Dec, 2018 2 commits
- remove debug code · cd034c8c
  Shining Sun authored Dec 18, 2018
  
  cd034c8c
- set learning rate for eager mode · 84266860
  Shining Sun authored Dec 18, 2018
  
  84266860
15 Dec, 2018 1 commit
- bug fix · b9e30b11
  Shining Sun authored Dec 15, 2018
  
  b9e30b11
14 Dec, 2018 5 commits
- bug fix · 6b673d03
  Shining Sun authored Dec 14, 2018
  
  6b673d03
- bug fixes · cf723306
  Shining Sun authored Dec 14, 2018
  
  cf723306
- first attempt to extract common parts of two main files · 3fd9c7fe
  Shining Sun authored Dec 14, 2018
  
  3fd9c7fe
- fix the dataset flag · eaf2bd1b
  Shining Sun authored Dec 14, 2018
  
  eaf2bd1b
- first attempt to merge the two main files · 84d3c62c
  Shining Sun authored Dec 14, 2018
  
  84d3c62c
12 Dec, 2018 1 commit
- allow specifying parse function · 3256c49f
  Priya Gupta authored Dec 11, 2018
  
  3256c49f
10 Dec, 2018 1 commit
- Modify to work with perfzero framework. · 2f2a04c7
  Toby Boyd authored Dec 10, 2018
  
  2f2a04c7
08 Dec, 2018 2 commits
- Add option to turn on eager in imagenet · c254938d
  Priya Gupta authored Dec 08, 2018
  
  c254938d
- Add a copy of Jing's ResNet50 model and use that in imagenet_main · aa1d6176
  Priya Gupta authored Dec 08, 2018
  
  aa1d6176
07 Dec, 2018 5 commits
- Add reporting of stats and turn off dist_strat. · 1b3c9ba6
  Toby Boyd authored Dec 07, 2018
  
  1b3c9ba6
- add flag to turn off dist_strat. · 46cc8af1
  Toby Boyd authored Dec 07, 2018
  
  46cc8af1
- Initial benchmark script idea. · 834cff94
  Toby Boyd authored Dec 07, 2018
  
  834cff94
- Add validation every epoch. Remove random seed setting. Change logging frequency · 3060b534
  Priya Gupta authored Dec 07, 2018
  
  3060b534
- Add validation per epoch · e17e64bb
  Priya Gupta authored Dec 07, 2018
  
  e17e64bb
05 Dec, 2018 1 commit
- Add option to run eager; make 1 GPU case run without DS · 59788849
  Priya Gupta authored Dec 05, 2018
  
  59788849
03 Dec, 2018 2 commits
- use v2 optimizer and dlr · b9d2b1bb
  Priya Gupta authored Dec 03, 2018
  
  b9d2b1bb
- Change LR schedule to adjust according to batch size · e932712b
  Priya Gupta authored Dec 03, 2018
  
  e932712b
30 Nov, 2018 6 commits
- Setting the initial learning rate. · 746a927c
  Amit Patankar authored Nov 30, 2018
  
  746a927c
- Setting the CIFAR parameters for the appropriate learning rate and num gpus. · 1cd1999e
  Amit Patankar authored Nov 30, 2018
  
  1cd1999e
- Updating the base learning rate to 3.2 · 3570932e
  Amit Patankar authored Nov 30, 2018
  
  3570932e
- Revert "Updating the learning_rate_fn for keras cifar." · abdc7f11
  Amit Patankar authored Nov 30, 2018
```
This reverts commit 649e0374.
```
  abdc7f11
- Updating the learning_rate_fn for keras cifar. · 649e0374
  Amit Patankar authored Nov 30, 2018
  
  649e0374
- Experimenting with cifar keras. · b2af31a9
  Amit Patankar authored Nov 30, 2018
  
  b2af31a9
27 Nov, 2018 3 commits
- Fix steps per epoch and tensorboard logging · 9f34ac99
  Priya Gupta authored Nov 27, 2018
  
  9f34ac99
- Actually add the keras resnet code · 950e1def
  Priya Gupta authored Nov 27, 2018
  
  950e1def
- Add keras resnet code · 31021959
  Priya Gupta authored Nov 27, 2018
  
  31021959
21 Nov, 2018 2 commits

worker_devices moving to ds.extended (#5775) · b2522f9b
josh11b authored Nov 20, 2018

b2522f9b

cross_tower_ops -> cross_device_ops (#5776) · 9a4848a2

josh11b authored Nov 20, 2018

We've deprecated the "tower" terminology in DistributionStrategy, so the "cross_tower_ops" argument is now "cross_device_ops", matching the current name of "AllReduceCrossDeviceOps".

9a4848a2

17 Nov, 2018 1 commit
- Parameter definition is in resnet_run_loop.py (#5770) · a6494752
  HelgeS authored Nov 17, 2018
  
  a6494752
14 Nov, 2018 1 commit

num_replicas -> num_replicas_in_sync (#5752) · 4d4eb85c

josh11b authored Nov 14, 2018

`num_replicas` is being replaced by `num_replicas_in_sync` and something that returns the number of steps running concurrently.

4d4eb85c

07 Nov, 2018 1 commit
- Fix PREPROC_HP_NUM_EVAL tag for MLPerf. (#5717) · d7ce21fa
  Reed authored Nov 07, 2018
```
This tag should match EVAL_HP_NUM_NEG.
```
  d7ce21fa