Welcome to MMPretrain’s documentation!¶
MMPretrain is a newly upgraded open-source framework for pre-training. It has set out to provide multiple powerful pre-trained backbones and support different pre-training strategies. MMPretrain originated from the famous open-source projects MMClassification and MMSelfSup, and is developed with many exiciting new features. The pre-training stage is essential for vision recognition currently. With the rich and strong pre-trained models, we are currently capable of improving various downstream vision tasks.
Our primary objective for the codebase is to become an easily accessible and user-friendly library and to streamline research and engineering. We detail the properties and design of MMPretrain across different sections.
Hands-on Roadmap of MMPretrain¶
To help users quickly utilize MMPretrain, we recommend following the hands-on roadmap we have created for the library:
For users who want to try MMPretrain, we suggest reading the GetStarted section for the environment setup.
For basic usage, we refer users to UserGuides for utilizing various algorithms to obtain the pre-trained models and evaluate their performance in downstream tasks.
For those who wish to customize their own algorithms, we provide AdvancedGuides that include hints and rules for modifying code.
To find your desired pre-trained models, users could check the ModelZoo, which features a summary of various backbones and pre-training methods and introfuction of different algorithms.
Additionally, we provide Analysis and Visualization tools to help diagnose algorithms.
Besides, if you have any other questions or concerns, please refer to the Notes section for potential answers.
We always welcome PRs and Issues for the betterment of MMPretrain.
Get Started
User Guides
Advanced Guides
Model Zoo
- Model Zoo Summary
- ArcFace
- BarlowTwins
- BEiT
- BEiTv2
- BLIP
- BLIP-2
- BYOL
- CAE
- ChineseCLIP
- CLIP
- Conformer
- ConvMixer
- ConvNeXt
- ConvNeXt V2
- CSPNet
- CSRA
- DaViT
- DeiT
- DeiT III: Revenge of the ViT
- DenseCL
- DenseNet
- DINOv2
- EdgeNeXt
- EfficientFormer
- EfficientNet
- EfficientNetV2
- EVA
- EVA-02
- Flamingo
- GLIP
- HiViT
- HorNet
- HRNet
- Inception V3
- iTPN
- LeViT
- LLaVA
- MAE
- MaskFeat
- MFF
- MILAN
- MiniGPT4
- MixMIM
- MLP-Mixer
- MobileNet V2
- MobileNet V3
- MobileOne
- MobileViT
- MoCoV2
- MoCoV3
- MViT V2
- OFA
- Otter
- PoolFormer
- RegNet
- RepLKNet
- RepMLP
- RepVGG
- Res2Net
- ResNet
- ResNeXt
- Reversible Vision Transformers
- RIFormer
- SAM
- SEResNet
- Shufflenet V1
- Shufflenet V2
- SimCLR
- SimMIM
- SimSiam
- SparK
- SwAV
- Swin-Transformer
- Swin-Transformer V2
- Tokens-to-Token ViT
- TinyViT
- Transformer in Transformer
- Twins
- Visual-Attention-Network
- VGG
- VIG
- Vision Transformer
- Wide-ResNet
- XCiT
Visualization
Analysis Tools
Deployment
Migration
API Reference
Notes
Device Support