Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Official Implementation of 'Fast AutoAugment' in PyTorch.
| Date | Stars |
|---|---|
| 2026-07-24 | 1605 |
| 2026-07-25 | 1605 |
| 2026-07-28 | 1605 |
| 2026-07-30 | 1605 |
| 2026-08-06 | 1605 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Fast AutoAugment **(Accepted at NeurIPS 2019)** Official [Fast AutoAugment](https://arxiv.org/abs/1905.00397) implementation in PyTorch. - Fast AutoAugment learns augmentation policies using a more efficient search strategy based on density matching. - Fast AutoAugment speeds up the search time by orders of magnitude while maintaining the comparable performances. <p align="center"> <img src="etc/search.jpg" height=350> </p> ## Results ### CIFAR-10 / 100 Search : **3.5 GPU Hours (1428x faster than AutoAugment)**, WResNet-40x2 on Reduced CIFAR-10 | Model(CIFAR-10) | Baseline | Cutout | AutoAugment | Fast AutoAugment<br/>(transfer/direct) | | |-------------------------|------------|------------|-------------|------------------|----| | Wide-ResNet-40-2 | 5.3 | 4.1 | 3.7 | 3.6 / 3.7 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar10_wresnet40x2_top1_3.52.pth) | | Wide-ResNet-28-10 | 3.9 | 3.1 | 2.6 | 2.7 / 2.7 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar10_wresnet28x10_top1.pth) | | Shake-Shake(26 2x32d) | 3.6 | 3.0 | 2.5 | 2.7 / 2.5 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar10_shake26_2x32d_top1_2.68.pth) | | Shake-Shake(26 2x96d) | 2.9 | 2.6 | 2.0 | 2.0 / 2.0 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar10_shake26_2x96d_top1_1.97.pth) | | Shake-Shake(26 2x112d) | 2.8 | 2.6 | 1.9 | 2.0 / 1.9 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar10_shake26_2x112d_top1_2.04.pth) | | PyramidNet+ShakeDrop | 2.7 | 2.3 | 1.5 | 1.8 / 1.7 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar10_pyramid272_top1_1.44.pth) | | Model(CIFAR-100) | Baseline | Cutout | AutoAugment | Fast AutoAugment<br/>(transfer/direct) | | |-----------------------|------------|------------|-------------|------------------|----| | Wide-ResNet-40-2 | 26.0 | 25.2 | 20.7 | 20.7 / 20.6 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar100_wresnet40x2_top1_20.43.pth) | | Wide-ResNet-28-10 | 18.8 | 18.4 | 17.1 | 17.3 / 17.3 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar100_wresnet28x10_top1_17.17.pth) | | Shake-Shake(26 2x96d) | 17.1 | 16.0 | 14.3 | 14.9 / 14.6 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar100_shake26_2x96d_top1_15.15.pth) | | PyramidNet+ShakeDrop | 14.0 | 12.2 | 10.7 | 11.9 / 11.7 | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/cifar100_pyramid272_top1_11.74.pth) | ### ImageNet Search : **450 GPU Hours (33x faster than AutoAugment)**, ResNet-50 on Reduced ImageNet | Model | Baseline | AutoAugment | Fast AutoAugment<br/>(Top1/Top5) | | |------------|------------|-------------|------------------|----| | ResNet-50 | 23.7 / 6.9 | 22.4 / 6.2 | **22.4 / 6.3** | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/imagenet_resnet50_top1_22.2.pth) | | ResNet-200 | 21.5 / 5.8 | 20.0 / 5.0 | **19.4 / 4.7** | [Download](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/imagenet_resnet200_top1_19.4.pth) | Notes * We evaluated resnet-50 and resnet-200 with resolution of 224 and 320, respectively. According to the original resnet paper, resnet 200 was tested with the resolution of 320. Also our resnet-200 baseline's performance was similar when we use the resolution. * But with recent our code clean-up and bugfixes, we've found that the baseline performs similar to the baseline even using 224x224. * When we use 224x224, resnet-200 performs **20.0 / 5.2**. Download link for the trained model is [here](https://arena.kakaocdn.net/brainrepo/fast-autoaugment/image
Excerpt of 7,920 characters
Read on GitHub
curtis.abcd · South Korea
56
2
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:891d82f9b4ba545f, topic:deep-learning, topic:pytorch
matched fp:891d82f9b4ba545f, topic:computer-vision, topic:image-classification