Unofficial implementation of the ImageNet, CIFAR 10 and SVHN Augmentation Policies learned by AutoAugment using pillow

Last update: Jan 02, 2023

Related tags

Overview

AutoAugment - Learning Augmentation Policies from Data

Unofficial implementation of the ImageNet, CIFAR10 and SVHN Augmentation Policies learned by AutoAugment, described in this Google AI Blogpost.

Update July 13th, 2018: Wrote a Blogpost about AutoAugment and Double Transfer Learning.

Tested with Python 3.6. Needs pillow>=5.0.0

Example

from autoaugment import ImageNetPolicy
image = PIL.Image.open(path)
policy = ImageNetPolicy()
transformed = policy(image)

To see examples of all operations and magnitudes applied to images, take a look at AutoAugment_Exploration.ipynb.

Example as a PyTorch Transform - ImageNet

from autoaugment import ImageNetPolicy
data = ImageFolder(rootdir, transform=transforms.Compose(
                        [transforms.RandomResizedCrop(224), 
                         transforms.RandomHorizontalFlip(), ImageNetPolicy(), 
                         transforms.ToTensor(), transforms.Normalize(...)]))
loader = DataLoader(data, ...)

Example as a PyTorch Transform - CIFAR10

from autoaugment import CIFAR10Policy
data = ImageFolder(rootdir, transform=transforms.Compose(
                        [transforms.RandomCrop(32, padding=4, fill=128), # fill parameter needs torchvision installed from source
                         transforms.RandomHorizontalFlip(), CIFAR10Policy(), 
			 transforms.ToTensor(), 
                         Cutout(n_holes=1, length=16), # (https://github.com/uoguelph-mlrg/Cutout/blob/master/util/cutout.py)
                         transforms.Normalize(...)]))
loader = DataLoader(data, ...)

Example as a PyTorch Transform - SVHN

from autoaugment import SVHNPolicy
data = ImageFolder(rootdir, transform=transforms.Compose(
                        [SVHNPolicy(), 
			 transforms.ToTensor(), 
                         Cutout(n_holes=1, length=20), # (https://github.com/uoguelph-mlrg/Cutout/blob/master/util/cutout.py)
                         transforms.Normalize(...)]))
loader = DataLoader(data, ...)

Results with AutoAugment

Generalizable Data Augmentations

Finally, we show that policies found on one task can generalize well across different models and datasets. For example, the policy found on ImageNet leads to significant improvements on a variety of FGVC datasets. Even on datasets for which fine-tuning weights pre-trained on ImageNet does not help significantly [26], e.g. Stanford Cars [27] and FGVC Aircraft [28], training with the ImageNet policy reduces test set error by 1.16% and 1.76%, respectively. This result suggests that transferring data augmentation policies offers an alternative method for transfer learning.

Unofficial implementation of the ImageNet, CIFAR 10 and SVHN Augmentation Policies learned by AutoAugment using pillow

Related tags

Overview

AutoAugment - Learning Augmentation Policies from Data

Tested with Python 3.6. Needs pillow>=5.0.0

Example

Example as a PyTorch Transform - ImageNet

Example as a PyTorch Transform - CIFAR10

Example as a PyTorch Transform - SVHN

Results with AutoAugment

Generalizable Data Augmentations

CIFAR 10

CIFAR 100

ImageNet

SVHN

Fine Grained Visual Classification Datasets

Owner

Philip Popien

ThunderGBM: Fast GBDTs and Random Forests on GPUs

Some simple programs built in Python: webcam with cv2 that detects eyes and face, with grayscale filter

A Python package for time series augmentation

PyTorch implementation of SIFT descriptor

NumQMBasic - A mini-course offered to Undergrad physics students

Pneumonia Detection using machine learning - with PyTorch

Official Implementation of "Third Time's the Charm? Image and Video Editing with StyleGAN3" https://arxiv.org/abs/2201.13433

Tracking code for the winner of track 1 in the MMP-Tracking Challenge at ICCV 2021 Workshop.

Ankou: Guiding Grey-box Fuzzing towards Combinatorial Difference

Offical implementation for "Trash or Treasure? An Interactive Dual-Stream Strategy for Single Image Reflection Separation".

Unofficial PyTorch reimplementation of the paper Swin Transformer V2: Scaling Up Capacity and Resolution

RL and distillation in CARLA using a factorized world model

Junction Tree Variational Autoencoder for Molecular Graph Generation (ICML 2018)

SegNet including indices pooling for Semantic Segmentation with tensorflow and keras

paper: Hyperspectral Remote Sensing Image Classification Using Deep Convolutional Capsule Network

Pytorch implementation of Each Part Matters: Local Patterns Facilitate Cross-view Geo-localization https://arxiv.org/abs/2008.11646

How the Deep Q-learning method works and discuss the new ideas that makes the algorithm work

Semi-supervised Representation Learning for Remote Sensing Image Classification Based on Generative Adversarial Networks

Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation

Deep Learning for Time Series Forecasting.