Commits · e455179a6a455628faae05b1673cb98bc53244ff · ModelZoo / ResNet50_tensorflow

"test/vscode:/vscode.git/clone" did not exist on "4f7886d1e07972b964547f318847a18a77f7aa97"

22 Mar, 2021 10 commits

Internal change · e455179a
Yeqing Li authored Mar 22, 2021
```
PiperOrigin-RevId: 364433023
```
e455179a
Internal change · f8eb66ea
A. Unique TensorFlower authored Mar 22, 2021
```
PiperOrigin-RevId: 364395434
```
f8eb66ea
Restore dummy computation function. · 7a02b5ce
Vighnesh Birodkar authored Mar 22, 2021
```
PiperOrigin-RevId: 364388006
```
7a02b5ce
Internal change · 2da86542
Poorva Potdar authored Mar 22, 2021
```
PiperOrigin-RevId: 364378436
```
2da86542
Add test for performance exporter · f9491103
Vighnesh Birodkar authored Mar 22, 2021
```
PiperOrigin-RevId: 364363163
```
f9491103

Add a new option "recreate_iterator_for_each_eval" into StandardEvaluator. · 0b7674b9

A. Unique TensorFlower authored Mar 22, 2021

In current implementation of StandardEvaluator, the iterator of evaluate dataset will go back (actually create a new iterator) to the beginning when evaluate() is called every time.

In some case, the iterator need to go ahead forever. So add the new option "recreate_iterator_for_each_eval" to control this behavior.

In some case, the iterator creation is time consuming. For example, careting an iterator of distributed dataset for multi worker in eager mode. So it's better to make a repeat dataset instead of recreating an iterator.

If "recreate_iterator_for_each_eval" is True, then the iterator will go back; Otherwise, keep going ahead.

PiperOrigin-RevId: 364356169

0b7674b9

Internal change. · 6a67bfdc
Vighnesh Birodkar authored Mar 22, 2021
```
PiperOrigin-RevId: 364349645
```
6a67bfdc
Update layer documentation. · 0c48b89f
Jeremiah Liu authored Mar 22, 2021
```
PiperOrigin-RevId: 364328905
```
0c48b89f
Implements `GaussianProcessClassificationHead`. · 572fafb8
Jeremiah Liu authored Mar 21, 2021
```
PiperOrigin-RevId: 364226289
```
572fafb8
Minor readability improvements. · 4785b025
Le Hou authored Mar 21, 2021
```
PiperOrigin-RevId: 364219111
```
4785b025

20 Mar, 2021 2 commits
- Adds spectral normalization and Gaussian process layers. · acf9fd92
  Jeremiah Liu authored Mar 20, 2021
```
PiperOrigin-RevId: 364109398
```
  acf9fd92
- remove calls to metrics.result() inside the train_step · cb17f22a
  A. Unique TensorFlower authored Mar 20, 2021
```
PiperOrigin-RevId: 364096705
```
  cb17f22a
19 Mar, 2021 8 commits
- Changing the default number of channels for lighter-weight hourglass backbones. · 0ea84d6d
  Ronny Votel authored Mar 19, 2021
```
PiperOrigin-RevId: 363975731
```
  0ea84d6d
- moved to tensorflow/examples · 5150eb3e
  Mark Daoust authored Mar 19, 2021
  
  5150eb3e
- Add Wide and deep by NVIDIA (#9780) · 8eccc9eb
  Pablo Ribalta Lorenzo authored Mar 19, 2021
  
  8eccc9eb
- Dataset utilities added. (#9796) · 7899ccd7
  Jekaterina Jaroslavceva authored Mar 19, 2021
  
  7899ccd7
- Update tflite support version in workspace (#9803) · e9ca9ffb
  thunderfyc authored Mar 19, 2021
  
  e9ca9ffb
- Update CODEOWNERS · 2657a7ca
  Jaeyoun Kim authored Mar 19, 2021
```
Change directory for seq_flow_lite
```
  2657a7ca
- Add image+box module in exporter. · 532b946c
  Vighnesh Birodkar authored Mar 18, 2021
```
PiperOrigin-RevId: 363794263
```
  532b946c
- [core] Only use runtime_options for training. · dfaf525e
  Frederick Liu authored Mar 18, 2021
```
PiperOrigin-RevId: 363782489
```
  dfaf525e
18 Mar, 2021 6 commits
- Rename pylint.sh to .github/scripts/pylint.sh · b6d1ec22
  Jaeyoun Kim authored Mar 18, 2021
```
Move to .github
```
  b6d1ec22
- Fixes typos and use string for endpoints key. · ec48284d
  Yeqing Li authored Mar 18, 2021
```
PiperOrigin-RevId: 363736548
```
  ec48284d
- Log the loss value in summary exporter. · 17263f30
  Vighnesh Birodkar authored Mar 18, 2021
```
PiperOrigin-RevId: 363700908
```
  17263f30
- Internal change · 99572269
  Dana Movshovitz-Attias authored Mar 18, 2021
```
PiperOrigin-RevId: 363673488
```
  99572269
- [core] Add tpu_enable_xla_dynamic_padder to RuntimeConfig. · 64d412cb
  Frederick Liu authored Mar 18, 2021
```
PiperOrigin-RevId: 363595377
```
  64d412cb
- Add a unit test to demonstrate how to load an encoder through yaml and checkpoint. · 739785bb
  Hongkun Yu authored Mar 17, 2021
```
PiperOrigin-RevId: 363593694
```
  739785bb
17 Mar, 2021 3 commits
- Internal change · bd39ad59
  Chen Chen authored Mar 17, 2021
```
PiperOrigin-RevId: 363529035
```
  bd39ad59
- Updated the target assigner and centernet meta arch such that it supports the · fe816dc2
  Yu-hui Chen authored Mar 17, 2021
```
use case which uses the keypoint location to determine the object center
location and heatmap.

PiperOrigin-RevId: 363522414
```
  fe816dc2
- Update TODOs for {Bert,Sentencepiece}Tokenizer.get_config(): · f84103f8
  A. Unique TensorFlower authored Mar 17, 2021
```
both are tracked in the same bug.

PiperOrigin-RevId: 363431367
```
  f84103f8
16 Mar, 2021 5 commits
- Add pylint wrapper extracted from TensorFlow CI. (#9793) · 3eedb03c
  Revan Sopher authored Mar 16, 2021
```
TensorFlow uses a subset of pylint functionality. Script will download the latest pylint rules from the main repo at runtime.

Requires `python3.8 -m pip install pylint==2.4.4`.
This will eventually be run on-PR in this repo, but first we'll have to fix the ~900 existing violations.
Script should be usable manually for now -- call it from within a subdirectory to avoid processing the entire repo.
```
  3eedb03c
- [nlp][translation] Consistently use float32 for layer norm for both encoder and decoder. · cebc9e90
  Frederick Liu authored Mar 16, 2021
```
PiperOrigin-RevId: 363233691
```
  cebc9e90
- Support arbitrary dimensions in stochastic depth. · 0537226c
  Dan Kondratyuk authored Mar 16, 2021
```
PiperOrigin-RevId: 363188927
```
  0537226c
- [nlp][translation] Remove seq2seq model _dtype argument and break transformer utils dependency. · 1180f37e
  Frederick Liu authored Mar 16, 2021
```
PiperOrigin-RevId: 363121226
```
  1180f37e
- Fix AdamWeightDecay: self.gradient_clip_norm should be used. · f06dc1a6
  Hongkun Yu authored Mar 15, 2021
```
PiperOrigin-RevId: 363069319
```
  f06dc1a6
15 Mar, 2021 1 commit
- Internal change · 69eb52cd
  A. Unique TensorFlower authored Mar 15, 2021
```
PiperOrigin-RevId: 362993979
```
  69eb52cd
14 Mar, 2021 2 commits
- Typo reported by https://github.com/tensorflow/models/pull/9672 · a4b513b1
  Hongkun Yu authored Mar 14, 2021
```
PiperOrigin-RevId: 362836016
```
  a4b513b1
- Removed the dependency of the keypoint task and object detection task in the · 0a17932d
  Yu-hui Chen authored Mar 14, 2021
```
postprocessing logic.

PiperOrigin-RevId: 362814113
```
  0a17932d
13 Mar, 2021 1 commit
- Adds tf.distribute.experimental.ParameterServerStrategy support to Orbit. · 2e77bb3e
  Yeqing Li authored Mar 13, 2021
```
PiperOrigin-RevId: 362729857
```
  2e77bb3e
12 Mar, 2021 2 commits
- Internal change · f7ea371e
  A. Unique TensorFlower authored Mar 12, 2021
```
PiperOrigin-RevId: 362613992
```
  f7ea371e
- Use backticks to denote code spans in nlp modeling docstrings · 94dd4911
  Chen Chen authored Mar 12, 2021
```
PiperOrigin-RevId: 362610475
```
  94dd4911