A friend of mine recently asked whether there was an easy way to introduce new levels to an embedding layer in Keras and only training the embedding layer for those new levels, i.e. keeping the weights for the existing indexes fixed.
Note that this is not necessarily a good …
I’ve recently bought a reMarkable 2 and am happy with it. Since teaching is still happening online for the most part at the moment, I was curious whether I could use it as a drawing tablet, share the screen with my machine, and then pick that up in e …
TL;DR: very likely not.
A student once asked me whether it would make sense to tune the probability cutoff (a.k.a. threshold) for random forests. Let’s explain first what we mean by that exactly. Assume a binary classification setting. Say we’ve trained a single decision tree …
In 2019, I wrote on the topic of discovering which features interact in a model agnostic manner, using an approach from a paper by Friedman and Popescu entitled Predictive Learning Via Rule Ensembles. I wanted to revisit the topic briefly and expand on it somewhat, given that scikit-learn has progressed …
It’s time to setup a new machine, so once again I’m jotting down a list of apps I generally install right away, slightly updated again.
I was going to spew this out as a Twitter “thread” as is the fashion these days, but I decided against it. Blogging is still cool. Especially once I realized this is not the first time I’ve done this…
… Meaning that I’m sitting at my machine and realize …
Like many others, I’m a huge fan of Jeremy Howard’s fast.ai courses. Even when having worked in the machine learning field for years, I still find the material to be packed full with interesting …
It’s time to setup a new laptop, so once again I’m jotting down a list of apps I normally install right away, slightly updated again.
Optional: