{"id":386,"date":"2016-01-13T03:06:24","date_gmt":"2016-01-13T01:06:24","guid":{"rendered":"http:\/\/log.or.cz\/?p=386"},"modified":"2016-01-13T18:00:12","modified_gmt":"2016-01-13T16:00:12","slug":"keras-for-binary-classification","status":"publish","type":"post","link":"https:\/\/log.or.cz\/?p=386","title":{"rendered":"Keras for Binary Classification"},"content":{"rendered":"

So I didn’t get around to seriously (besides running a few examples) play with Keras<\/a><\/b> (a powerful library for building fully-differentiable machine learning models aka neural networks<\/b>) – until now. And I have been a bit surprised about how tricky it actually was for me to get a simple task running, despite (or maybe because of) all the docs available already.<\/p>\n

The thing is, many of the “basic examples” gloss over exactly how the inputs and mainly outputs look like, and that’s important<\/em>. Especially since for me, the archetypal simplest machine learning problem consists of binary classification<\/b>, but in Keras the canonical task is categorical classification<\/b>. Only after fumbling around for a few hours, I have realized this fundamental rift.<\/p>\n

The examples (besides LSTM sequence classification) silently assume that you want to classify to categories (e.g. to predict words etc.), not do a binary 1\/0 classification. The consequences are that if you naively copy the example MLP at first, before learning to think about it, your model will never learn anything and to add insult to injury, always show the accuracy as 1.0.<\/p>\n

So, there are a few important things you need to do to perform binary classification:<\/p>\n