Hang Zhang

The model access has been fixed. Could you reinstall the newest version of the code?

Did you reinstall the library?

We have provided detectron2 wrapper. https://github.com/zhanghang1989/ResNeSt/tree/master/d2 Everything works the same with detectron2. You can easily add your own data through registry.

``` from resnest.d2 import add_resnest_config add_resnest_config(cfg) ``` https://github.com/zhanghang1989/ResNeSt/blob/master/d2/train_net.py#L123

你好，我们模型是使用 mxnet 训练的，可以使用这个命令来 reproduce 文中的结果： https://github.com/zhanghang1989/ResNeSt/tree/master/scripts/gluon#training-command

The model definition can be found at https://github.com/zhanghang1989/ResNeSt/pull/29 There is no pytorch pretrained weights, but there is a mxnet version https://cv.gluon.ai/model_zoo/classification.html#resnest

你好，关于 RFConv 可以看这里 https://github.com/zhanghang1989/RFConv 比标准的 conv 会好一些，但是需要重新训练模型

I think your pytorch version is out-of-date. https://pytorch.org/docs/master/hub.html?highlight=load_state_dict_from_url#torch.hub.load_state_dict_from_url You may update pytorch or remove that argument.

Thanks for the quick reply! I used different training HPs `learning rate: 0.2, batch size: 512, weight decay: 1e-4` When switching back to paper default: `learning rate: 0.8, batch size:...

Thanks @uvipen Could you the results in your experiments? I saw the README is still pending.