Commits · 1e1353da3ab09b93371b0cb8ba7cc0c26b9790c9 · ModelZoo / ResNet50_tensorflow

12 Apr, 2021 1 commit
- Internal change · 1e1353da
  Reed Wanderman-Milne authored Apr 12, 2021
```
PiperOrigin-RevId: 368067415
```
  1e1353da
05 Apr, 2021 1 commit

Use nonexperimental LSO API in base_task.py. · 2ca3440e

Reed Wanderman-Milne authored Apr 05, 2021

This shouldn't break any official models, since I changed all LossScaleOptimizer isinstance checks to use the nonexperimental version (the experimental LSO subclasses the nonexperimental LSO, so changing isinstance checks in this way is always safe).

PiperOrigin-RevId: 366891847

2ca3440e

23 Mar, 2021 1 commit
- Internal change · d5afcc72
  Yeqing Li authored Mar 22, 2021
```
PiperOrigin-RevId: 364488384
```
  d5afcc72
03 Mar, 2021 2 commits

Pass use_experimental_api=True to mixed precision functions. · 5b9f6e2f

Reed Wanderman-Milne authored Mar 03, 2021

The default is True, but I plan on changing it to False soon. After that, I plan on removing the argument and never using the experimental API.

PiperOrigin-RevId: 360724698

5b9f6e2f

Pass use_experimental_api=True to mixed precision functions. · 4ab6f1b1

Reed Wanderman-Milne authored Mar 03, 2021

The default is True, but I plan on changing it to False soon. After that, I plan on removing the argument and never using the experimental API.

PiperOrigin-RevId: 360724698

4ab6f1b1

01 Mar, 2021 2 commits
- [core] Always call build_and_prepare since it is no op if exist. · c2793c1b
  Frederick Liu authored Mar 01, 2021
```
https://www.tensorflow.org/datasets/api_docs/python/tfds/load

PiperOrigin-RevId: 360237231
```
  c2793c1b
- [core] Always call build_and_prepare since it is no op if exist. · d8a0d379
  Frederick Liu authored Mar 01, 2021
```
https://www.tensorflow.org/datasets/api_docs/python/tfds/load

PiperOrigin-RevId: 360237231
```
  d8a0d379
27 Jan, 2021 1 commit

YOLO Family: Data Loaders (#9493) · 83b992c5

Akhil Chinnakotla authored Jan 27, 2021



* Data Loading

* data loading

* data loading

* data loading

* Bug fixes

* p

* documentation

* preprocessing_ops

* Dataloader PR Clean-Up

* First Set Of Code Review Fixes

* deleting imagenet

* Testing Functions

* Preserve aspect ratio
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* Put back the testing functions
Co-Authored-By: Akhil Chinnakotla <The-Indian-Chinna@users.noreply.github.com>

* Lint
Co-Authored-By: Akhil Chinnakotla <The-Indian-Chinna@users.noreply.github.com>

* Add forgotten build_grided_gt function
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* Add postprocessing function back
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* Change set to list
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* Fix small bugs
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* Add test case for training
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* Rename utils to ops
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* Remove pct_rand
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* More correct documentationin box_ops

* Remove junk files

* added tests for preprocessing_ops

* Add more descriptive comments

* Do not hardcode randomscale

* Remove useless BoxOps
Co-Authored-By: Vishnu Banna <vishnubanna@users.noreply.github.com>

* added test for box_ops

* Merge branch 'dataloaders_pr' of https://github.com/PurdueCAM2Project/tf-models

 into dataloaders_pr

* Updated Docstring

* Lint
Co-Authored-By: Akhil Chinnakotla <The-Indian-Chinna@users.noreply.github.com>
Co-authored-by: anivegesana <anirudh.vegesana@gmail.com>
Co-authored-by: Tyan3001 <yan262@purdue.edu>
Co-authored-by: Vishnu Banna <vishnubanna@users.noreply.github.com>
Co-authored-by: Akhil Chinnakotla <The-Indian-Chinna@users.noreply.github.com>

83b992c5

22 Jan, 2021 2 commits
- Internal change · a2201bb8
  Le Hou authored Jan 22, 2021
```
PiperOrigin-RevId: 353327814
```
  a2201bb8
- Internal change · 39abb95d
  Le Hou authored Jan 22, 2021
```
PiperOrigin-RevId: 353327814
```
  39abb95d
16 Nov, 2020 2 commits
- Limit Markdown file to 80 chars per line · 84df9351
  Anirudh Vegesana authored Nov 16, 2020
  
  84df9351
- Remove _use_bias parameter in DarkConv · dc2af579
  anivegesana authored Nov 16, 2020
  
  dc2af579
13 Nov, 2020 1 commit
- Remove dead files · 08ec7955
  anivegesana authored Nov 13, 2020
  
  08ec7955
08 Nov, 2020 2 commits
- cleaner grouping · 230732b2
  vishnubanna authored Nov 08, 2020
  
  230732b2
- cleaner grouping · a7f56602
  vishnubanna authored Nov 08, 2020
  
  a7f56602
07 Nov, 2020 5 commits
- ready for review · d5747aac
  vishnubanna authored Nov 07, 2020
  
  d5747aac
- pr comment to explain decoder · 31c3ab9e
  vishnubanna authored Nov 07, 2020
  
  31c3ab9e
- pr comment to explain decoder · 4af9988d
  vishnubanna authored Nov 07, 2020
  
  4af9988d
- pr comment to explain decoder · 1d658f4f
  vishnubanna authored Nov 07, 2020
  
  1d658f4f
- Kwargs update · a800e9b1
  The-Indian-Chinna authored Nov 07, 2020
  
  a800e9b1
06 Nov, 2020 3 commits
- activation function fix · dd5cc386
  vishnubanna authored Nov 06, 2020
  
  dd5cc386
- fixed activation functions · 15a302ed
  vishnubanna authored Nov 06, 2020
  
  15a302ed
- Turn DarkTiny from building block into stack · 3dc083ab
  anivegesana authored Nov 06, 2020
  
  3dc083ab
02 Nov, 2020 2 commits
- tfds csp · 5e85759a
  vishnubanna authored Nov 01, 2020
  
  5e85759a
- tfds csp · 0eebd4ba
  vishnubanna authored Nov 01, 2020
  
  0eebd4ba
01 Nov, 2020 12 commits
- tfds csp · 669807e6
  vishnubanna authored Nov 01, 2020
  
  669807e6
- Fix typos · da0d0d27
  anivegesana authored Nov 01, 2020
  
  da0d0d27
- load tests into one file and inherit from the imagenet task and the... · 47545935
  vishnubanna authored Nov 01, 2020
```
load tests into one file and inherit from the imagenet task and the functioning tfds decoder for imagenet
```
  47545935
- currently training darknet · 48b412c3
  Vishnu Banna authored Nov 01, 2020
  
  48b412c3
- darknet training on imagenet2012 with a new tfds decoder that works with the... · 6d3cfef4
  Vishnu Banna authored Nov 01, 2020
```
darknet training on imagenet2012 with a new tfds decoder that works with the existing classification example parser
```
  6d3cfef4
- tfds decoder test · 50dceb71
  vishnubanna authored Nov 01, 2020
  
  50dceb71
- tfds decoder test · 4ae6c23b
  vishnubanna authored Nov 01, 2020
  
  4ae6c23b
- undo all changes in this commit · d791ef44
  Vishnu Banna authored Nov 01, 2020
  
  d791ef44
- debug value · 9d58e701
  vishnubanna authored Nov 01, 2020
  
  9d58e701
- debug value · ff8d91aa
  vishnubanna authored Nov 01, 2020
  
  ff8d91aa
- rm change to orbit in final, testing backbone training · f1e8e671
  Vishnu Banna authored Nov 01, 2020
  
  f1e8e671
- imageclassification decoder for tfds · eeb371d4
  vishnubanna authored Oct 31, 2020
  
  eeb371d4
28 Oct, 2020 3 commits
- config fix · ded0e258
  vishnubanna authored Oct 28, 2020
  
  ded0e258
- YAML clean up · 6831c4d4
  The-Indian-Chinna authored Oct 28, 2020
  
  6831c4d4
- layers merging into one file · 5ec8552c
  The-Indian-Chinna authored Oct 28, 2020
  
  5ec8552c