Commits · 9485aa1dcf2c2e1ebaede0af0283796dcfb01911 · ModelZoo / ResNet50_tensorflow

30 Mar, 2023 1 commit
- initial files for ResNet50 · 3d61d6b3
  qianyj authored Mar 30, 2023
  
  3d61d6b3
19 Nov, 2021 1 commit
- Internal change · d1ed379e
  Hongkun Yu authored Nov 19, 2021
```
PiperOrigin-RevId: 411069186
```
  d1ed379e
15 Nov, 2021 1 commit
- Internal change · 65a967e9
  Hongkun Yu authored Nov 15, 2021
```
PiperOrigin-RevId: 410013894
```
  65a967e9
23 Jun, 2021 2 commits

Return default strategy from get_distribution_strategy when given "off". · 9f9d07e9

Reed Wanderman-Milne authored Jun 22, 2021

Before, it returned None. But almost every use of get_distribution_strategy() assumes an actual strategy is returned and crashes when None is returned. Returning the default strategy fixes these issues and is equivalent to using no strategy, as the default strategy is always in effect when no other strategy is used.

PiperOrigin-RevId: 380951055

9f9d07e9

Return default strategy from get_distribution_strategy when given "off". · 6dc4ae73

Reed Wanderman-Milne authored Jun 22, 2021

Before, it returned None. But almost every use of get_distribution_strategy() assumes an actual strategy is returned and crashes when None is returned. Returning the default strategy fixes these issues and is equivalent to using no strategy, as the default strategy is always in effect when no other strategy is used.

PiperOrigin-RevId: 380951055

6dc4ae73

25 Feb, 2021 2 commits

Fix issue where distribution_strategy="off" did not work. · 4864e795

Reed Wanderman-Milne authored Feb 24, 2021

Also improve error message when distribution=off is specified without properly quoting "off"

PiperOrigin-RevId: 359395329

4864e795

Fix issue where distribution_strategy="off" did not work. · 711bd150

Reed Wanderman-Milne authored Feb 24, 2021

Also improve error message when distribution=off is specified without properly quoting "off"

PiperOrigin-RevId: 359395329

711bd150

05 Jan, 2021 2 commits
- #KerasNLP Update TransformerEncoderBlock to support Q, KV as two input streams. · 97807e63
  Hongkun Yu authored Jan 05, 2021
```
PiperOrigin-RevId: 350170448
```
  97807e63
- #KerasNLP Update TransformerEncoderBlock to support Q, KV as two input streams. · 86df41f7
  Hongkun Yu authored Jan 05, 2021
```
PiperOrigin-RevId: 350170448
```
  86df41f7
29 Dec, 2020 2 commits
- Internal change · 7786b741
  Hongkun Yu authored Dec 28, 2020
```
PiperOrigin-RevId: 349347368
```
  7786b741
- Internal change · 002959e8
  Hongkun Yu authored Dec 28, 2020
```
PiperOrigin-RevId: 349347368
```
  002959e8
13 Sep, 2020 2 commits
- [Clean up] Consolidate distribution utils. · 420fd1cf
  Hongkun Yu authored Sep 12, 2020
```
PiperOrigin-RevId: 331359058
```
  420fd1cf
- [Clean up] Consolidate distribution utils. · 33a4c207
  Hongkun Yu authored Sep 12, 2020
```
PiperOrigin-RevId: 331359058
```
  33a4c207
12 Aug, 2020 2 commits
- Internal change · 999fae62
  Hongkun Yu authored Aug 12, 2020
```
PiperOrigin-RevId: 326286926
```
  999fae62
- Internal change · 88253ce5
  Hongkun Yu authored Aug 12, 2020
```
PiperOrigin-RevId: 326286926
```
  88253ce5
30 Mar, 2020 1 commit
- Move a R1 specific util function from common utils to R1 models. · fc02382c
  Hongkun Yu authored Mar 30, 2020
```
PiperOrigin-RevId: 303767122
```
  fc02382c
13 Feb, 2020 1 commit
- [Refactor] TF models: move all contents of transformer to nlp/transformer · 7f926353
  Hongkun Yu authored Feb 13, 2020
```
PiperOrigin-RevId: 294997928
```
  7f926353
29 Apr, 2019 1 commit

Replace per_device with per_replica and PerDevice with PerReplica, because the... · b00783d7

Igor authored Apr 29, 2019

Replace per_device with per_replica and PerDevice with PerReplica, because the PerDevice concept was renamed and doesn't exist anymore. (#6693)

* Replace per_device with per_replica and PerDevice with PerReplica, because the PerReplica concept was renamed and doesn't exist anymore.

b00783d7

28 Feb, 2019 1 commit
- Updating stale DistributionStrategy test. (#6281) · 4b566d4e
  Tayo Oguntebi authored Feb 28, 2019
  
  4b566d4e
13 Feb, 2019 1 commit

Add a flag to specify distribution strategies. (#6185) · 79b57a3f

Yuefeng Zhou authored Feb 12, 2019

* Add a flag to specify distribution strategies.

* Fix a small error.

* Address comments.

* Address comments.

* Fix typos.

79b57a3f

21 Nov, 2018 1 commit
- worker_devices moving to ds.extended (#5775) · b2522f9b
  josh11b authored Nov 20, 2018
  
  b2522f9b
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

24 Oct, 2018 1 commit
- Rename "num_towers" to "num_replicas" (#5599) · d7676c1c
  josh11b authored Oct 24, 2018
```
To match new terminology in DistributionStrategy.
```
  d7676c1c
18 Oct, 2018 1 commit

Drop references to `is_single_tower`. · b860839f

josh11b authored Oct 17, 2018

Since we plan on deleting this method, it is only used in distribution_utils_test.py.

b860839f

12 Jun, 2018 1 commit

Transformer multi gpu, remove multi_gpu flag, distribution helper functions (#4457) · 29c9f985

Katherine Wu authored Jun 12, 2018

* Add DistributionStrategy to transformer model

* add num_gpu flag

* Calculate per device batch size for transformer

* remove reference to flags_core

* Add synthetic data option to transformer

* fix typo

* add import back in

* Use hierarchical copy

* address PR comments

* lint

* fix spaces

* group train op together to fix single GPU error

* Fix translate bug (sorted_keys is a dict, not a list)

* Change params to a default dict (translate.py was throwing errors because params didn't have the TPU parameters.)

* Address PR comments. Removed multi gpu flag + more

* fix lint

* fix more lints

* add todo for Synthetic dataset

* Update docs

29c9f985