Shypd.aiThe largest open index of AI toolsOpenMixup is an open-source toolbox for supervised, semi-supervised, and self-supervised visual representation learning. It provides popular backbones, mixup methods, and algorithms for image classification and self-supervised pre-training.
Trending
About
OpenMixup is an open-source, PyTorch-based toolbox designed for supervised, semi-supervised, and self-supervised visual representation learning, with a particular focus on mixup-related methods. It offers a modular design, similar to OpenMMLab projects, allowing users to easily build customized models by combining various components. The toolbox includes popular backbones, a wide array of mixup data augmentations, and algorithms for both image classification (CNN & Transformer) and self-supervised pre-training (contrastive and autoregressive). It supports standard benchmarks for image classification, mixup classification, and self-supervised evaluation, with smooth integration for downstream tasks like object detection and segmentation. OpenMixup is actively updated to incorporate state-of-the-art methods and supports PyTorch versions 1.8 and higher.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
