Iman Mirzadeh

mobilenet, shufflenet, .... are already implemented and pretrained. they are available via torchvision models: https://pytorch.org/docs/stable/torchvision/models.html

Hi, Thank you for your comments! Let me explain a few things: 1- Seed Optimization This is done not to optimize for better seeds and is done to have consistent...

Yes. You are right. I'll fix it soon. Thanks :)

Hi, Sorry for the late reply. I don't have the parameters for Imagenet. That experiment was implemented by my co-authors at DeepMind. For CIFAR-100, can you please try [lambda_student =...

Hi. It seems that there is an error in train.py file. I will check it out and get back to you. Thank you for the notice :)

I'm not sure but I think the value 'T' here is None. what is the value of your config object?

Yes, one fix could be to set the config for yourself. Here => [https://github.com/imirzadeh/Teacher-Assistant-Knowledge-Distillation/blob/master/train.py#L168](https://github.com/imirzadeh/Teacher-Assistant-Knowledge-Distillation/blob/master/train.py#L168 ), define something like: `config = { "lambda_student": 0.5, "T_student": 5, "seed": 20 }` rather than:...

The way dark/light mode is chosen is based on this file: [https://github.com/wowchemy/wowchemy-hugo-modules/blob/main/wowchemy/assets/js/wowchemy-theming.js](https://github.com/wowchemy/wowchemy-hugo-modules/blob/main/wowchemy/assets/js/wowchemy-theming.js) More specifically, this function ```javascript function getThemeMode() { return parseInt(localStorage.getItem('wcTheme') || 2); } ``` By default, it returns...

I fixed the problem by changing `encrypt: true` in configuration.js file, also providing server ssl keys when running aria2c! ```--rpc-certificate, --rpc-private-key and --rpc-secure```

Hi, The problem is the module is load with dataparallel activated and you are trying to load it without data parallel. That's why there's an extra module at the beginning...