( Image credit: Albumentations )
In our implementation, we have designed a search space where a policy consists of many sub-policies, one of which is randomly chosen for each image in each mini-batch.
SOTA for Image Classification on SVHN
On LibriSpeech, we achieve 6. 8% WER on test-other without the use of a language model, and 5. 8% WER with shallow fusion with a language model.
#2 best model for Speech Recognition on Hub5'00 SwitchBoard
We provide examples of image augmentations for different computer vision tasks and show that Albumentations is faster than other commonly used image augmentation tools on the most of commonly used image transformations.
Additionally, due to the separate search phase, these approaches are unable to adjust the regularization strength based on model or dataset size.
SOTA for Image Classification on CIFAR-10 (Top 1 Accuracy metric )
In this paper, we introduce Random Erasing, a new data augmentation method for training the convolutional neural network (CNN).
#2 best model for Image Classification on Fashion-MNIST
We demonstrate that by making subtle but important changes to the model architecture and the learning rate schedule, fine-tuning image features, and adding data augmentation, we can significantly improve the performance of the up-down model on VQA v2. 0 dataset -- from 65. 67% to 70. 22%.
#2 best model for Visual Question Answering on VQA v2
During the learning of the student, we inject noise such as dropout, stochastic depth, and data augmentation via RandAugment to the student so that the student generalizes better than the teacher.
#2 best model for Image Classification on ImageNet (using extra training data)
Importantly, the best policy found on COCO may be transferred unchanged to other detection datasets and models to improve predictive accuracy.
#9 best model for Object Detection on COCO test-dev