Omiita

ViT-CIFAR. PyTorch implementation for Vision Transformer[Dosovitskiy, A.(ICLR'21)] modified to obtain over 90% accuracy FROM SCRATCH on CIFAR-10 with small number of parameters (= 6.3M, originally ViT-B has 86M).

MLP-Mixer-CIFAR. PyTorch implementation of Mixer-nano (#parameters is 0.67M, originally Mixer-S/16 has 18M) with 90.83 % acc. on CIFAR-10. Training from scratch.

japanese-lora-llm. A collection of Japanese LoRA-tuned LLMs.

TransGAN-PyTorch. (Ongoing) Unofficial re-implementation of TransGAN[Jiang, Y.(2021)].

ShuffleChannelLayer. Novel regularization by shuffling channels of feature maps.

image-classification-pytorch. A collection of image classification models along with results for CIFAR-10/100.