Roadmap of MMClassification

open-mmlab / mmpretrain

OpenMMLab Pre-training Toolbox and Benchmark

https://mmpretrain.readthedocs.io/en/latest/

Apache License 2.0

3.29k stars 1.03k forks source link

Roadmap of MMClassification #4

Open hellock opened 3 years ago

hellock commented 3 years ago

We keep this issue open to collect feature requests from users and hear your voice. Our monthly release plan is also available here.

You can either:

Suggest a new feature by leaving a comment.
Vote for a feature request with 👍 or be against with 👎. (Remember that developers are busy and cannot respond to all feature requests, so vote for your most favorable one!)
Tell us that you would like to help implement one of the features in the list or review the PRs. (This is the greatest things to hear about!)

GoldenAbel commented 3 years ago

warm up and more pretrained models

LeeJZh commented 3 years ago

tricks like mix-up, auto-augmentation, and etc..

Tshzzz commented 3 years ago

convert model to onnx.

SilvioGiancola commented 3 years ago

EfficientNet, that could serve as backbone for mmdetection as well

Ironteen commented 3 years ago

some Nerual Architecture Search algorithms which could be implement on one GPU, like ENAS

soldier828 commented 3 years ago

multi-brach for multi-task learning

LeeJZh commented 3 years ago

convert model to onnx.

I think pytorch 1.5 supports it. torch.onnx

GoldenAbel commented 3 years ago

darknet networks and data aug, like darknet53, mosaic, cutmix, mixup

myopengit commented 3 years ago

benchmark on cifar10 and cifar100

zhongqiu1245 commented 3 years ago

Could you support HRNet?

ousinkou commented 3 years ago

Multi-label classification

Kongsea commented 3 years ago

More types of loss supported, such as center loss, amsoftmax loss, and so on.

zhongqiu1245 commented 3 years ago

focal loss pls！

hiyyg commented 3 years ago

vovnet-v2, it is faster and more accurate than resnet in my experience.

hiyyg commented 3 years ago

HS-ResNet: Hierarchical-Split Block on Convolutional Neural Network https://arxiv.org/abs/2010.07621 There is a reference repo: https://github.com/bobo0810/HS-ResNet, but no pretrained models are provided. It would be nice if mmcls could include this and provide some pretrained models.