Karan Dwivedi

@bkj Did you manage to reproduce the results? I too implemented from scratch but am getting around 82% accuracy.

Thanks! I am doing the same but getting ~82%. Have you open-sourced your code (or can you please share your code)?

Nope. I couldn't run this. However, the authors have released their official code (link to it is in the README on this repo). ᐧ On Mon, Apr 30, 2018 at...

Are you using pytorch 0.4? ```python import torch print(torch.__version__) ```

Yes

You can see [this paper](https://arxiv.org/pdf/1605.07648.pdf) (Sec 3.1 specifically)

"child_out_filters" -> Number of channels of the convolution layers "child_num_branches", -> Number of options in ENAS (here it is fixed at 5) "child_num_cell_layers" -> Can't find this in the code.....

> Hence, up until now, this has been left to the application to figure this out and restart the modem if needed. Can someone please point to any guidance on...

Also, why not do avg pool with kernel size 2 instead? Is there a benefit of this approach?