authman

icon indicating an email [email protected]

icon location Irving, TX icon location Programming, machine learning engineering, healthtech.

Results 5 comments of authman

The layer before the auxiliary loss layer

I've tried with {ProjectionConv, BN, Relu, Dropout, Conv2d->NumClasses} as an aux branch. The results were 'meh'.

[Feature request] Add the Noisy Students pretrained weights

I tried porting it too. Most of the errors I got were patched using tf.compat.v1 wherever it complained—except tf.train.Saver(), which apparently doesn't exist. EDIT: was able to get around that...

[Feature request] Add the Noisy Students pretrained weights

I've tested it and it works great. I've also converted b4 if anyone is interested.

[Feature request] Add the Noisy Students pretrained weights

https://drive.google.com/file/d/1hORyc4b8_QPs1O67br5T_a9XeB92ICkF/view?usp=sharing B4 noisy student pth.

[Feature request] Add the Noisy Students pretrained weights

You guys can easily convert the files yourselves. There is a script in this very repo: https://github.com/lukemelas/EfficientNet-PyTorch/tree/master/tf_to_pytorch Use it to download tf weights and convert to .pth. For example, I...