https://github.com/ayulockin/dataaugmentationtf
Implementation of modern data augmentation techniques in TensorFlow 2.x to be used in your training pipeline.
https://github.com/ayulockin/dataaugmentationtf
augmix computer-vision cutmix cutout data-augmentation deeplearning keras mixup tensorflow2
Last synced: 3 months ago
JSON representation
Implementation of modern data augmentation techniques in TensorFlow 2.x to be used in your training pipeline.
- Host: GitHub
- URL: https://github.com/ayulockin/dataaugmentationtf
- Owner: ayulockin
- Created: 2020-05-21T12:11:30.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2020-07-03T07:56:21.000Z (about 5 years ago)
- Last Synced: 2025-04-30T18:06:12.553Z (3 months ago)
- Topics: augmix, computer-vision, cutmix, cutout, data-augmentation, deeplearning, keras, mixup, tensorflow2
- Language: Jupyter Notebook
- Homepage:
- Size: 12.8 MB
- Stars: 34
- Watchers: 1
- Forks: 12
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Modern Data Augmentation Techniques for Computer Vision
## About
Implementation of various data augmentation techniques in TensorFlow 2.x. They can be easily used in your training pipeline. This repository contain the supplementary notebooks for the [Modern Data Augmentation Techniques for Computer Vision](https://app.wandb.ai/authors/tfaugmentation/reports/Modern-Data-Augmentation-Techniques-for-Computer-Vision--VmlldzoxNDA2NTc)(Weights and Biases) report.
## Techniques Covered
* [Cutout](https://github.com/ayulockin/DataAugmentationTF/blob/master/CIFAR_10_with_Cutout_Augmentation.ipynb)
* [Mixup](https://github.com/ayulockin/DataAugmentationTF/blob/master/CIFAR_10_with_Mixup_Augmentation.ipynb)
* [CutMix](https://github.com/ayulockin/DataAugmentationTF/blob/master/CIFAR_10_with_CutMix_Augmentation.ipynb)
* [Augmix](https://github.com/ayulockin/DataAugmentationTF/blob/master/Cifar_10_with_AugMix_Augmentation.ipynb)
**Note**: Cutout, Mixup and CutMix are implememted in `tf.data` and can be found in the linked colab notebooks. I am using TensorFlow 2.x implementation of AugMix by [Aakash Nain](https://twitter.com/A_K_Nain?s=09). His repo can be found [here](https://github.com/AakashKumarNain/AugMix_TF2). The [fork](https://github.com/ayulockin/AugMix_TF2) of this repo contains Weights and Biases integration and some additional command like arguments for more control.
## Result
Check out the linked [report](https://app.wandb.ai/authors/tfaugmentation/reports/Modern-Data-Augmentation-Techniques-for-Computer-Vision--VmlldzoxNDA2NTc) for:
* The comparative study of these augmentation techniques.
* Augmentation implementations.
* Evaluation of these augmentation techniques against [Cifar-10-C dataset](https://zenodo.org/record/2535967).
## Model Used
