Hacker News new | comments | show | ask | jobs | submit login

> did you first download and use the pretrained net of an different classification task and use that as a default ?

Yes, but that did not give me accuracy enough, so now I train from scratch. I had hoped to save having to train the conv layers.

> also, did you use batch normalization?


> also, did you try ResNets?


> you probably don't care at this point, but all of this would __massively__ decrease training time

Oh, I care all right :) I'm re-training the net every evening (it's running right now) after adding another batch of training images.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact