Fast-PBA_AutoAugment

KO | EN

gitlite — search

#javascript #python #typescript #machine-learning #rust #deep-learning #linux #docker #nodejs #macos #windows #cli #mcp #open-source

Fast-PBA_AutoAugment

(★ 9)

Using Fast-AutoAugment and Population Based Augmentation to train deep neural networks on CIFAR-100 (w/ Pytorch)..

Fast-PBA_AutoAugment Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Comparison of Fast AutoAugment and Population Based Augmentation

The code is an adaptation of the Official [Fast AutoAugment](https://arxiv.org/abs/1905.00397) implementation in
PyTorch available [here](https://github.com/kakaobrain/fast-autoaugment/).

Fast AutoAugment is a simplified version of the well known [AutoAugment](https://arxiv.org/abs/1805.09501) paper of
Google Brain.

- 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.

The code is also modified so that it can use the augmentation schedule policiy obtained via the official implementation of
the [Population Based Augmentation](https://arxiv.org/abs/1905.05393) available [here](https://github.com/arcelien/pba).
Population Base Augmentation is a method that learns a data-augmentation schedule policy and applied during training.
The underlying intuition being that at the beginning, there is no need to perform a very elaborated data-augmentation.
However, the more the neural network learns, the better the data-augmentation should be.
Data-augmentation schedule policies can be found in the folder *schedules*.

## Table Results

| Model(CIFAR-100)      | Baseline   | Cutout     | AutoAugment | Fast AutoAugment<br/>(transfer/direct) |PBA|
|-----------------------|------------|------------|-------------|------------------|------------------|
| PyramidNet+ShakeDrop  | 14.0       | 12.2       | 10.7        | 11.7      | 11.4     |

## Run

### Prerequisite:
Download CIFAR-100 dataset.

### How to run:
```
cd /usr/share/bind_mount/scripts/MicroNet/Training
pip install -r requirements.txt
export PYTHONPATH="$PYTHONPATH:/usr/share/bind_mount/scripts/MicroNet/Training"
cd FastAutoAugment
horovodrun -np 4 --cache-capacity 2048 python train.py -c confs/pyramid272a200b.yaml --tag pyramidnet_cifar100 --horovod
```

The argument *np* allows you to choose the number of GPUs you want to use (here 4).

Please see the *confs/pyramid272a200b.yaml* file to get more insights on the parameters used for the training process.
You may notice an argument named *aug*. It allows you to choose which data augmentation policy you want to use.

The checkpoint weights are regularly saved (every 10 epochs) and put in the directory */app/results/checkpoints*.

Fast-PBA_AutoAugment

Fast-PBA_AutoAugment Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Fast-PBA_AutoAugment

Fast-PBA_AutoAugment Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README