• 8 Posts
  • 41 Comments
Joined 1 year ago
cake
Cake day: June 11th, 2023

help-circle









  • I don’t know how much content there is to share, so you might be overflowing with things to talk about every week. But I fear that doing so much effort weekly could be unsustainable. I would suggest, or hope you consider, a less freqent blog/podcast, like every other week. I think this would be more modest, and easier to maintain. You wouldn’t need to change to title of the blog, either.

    Anyway, this is an exciting project and I’m thankful for your work.










  • For more “traditional” or “statistical” modeling (not NN) 100% start with sklearn. It has a plethora of algorithms, and their docs read like a book. You can learn a whole bunch of new methods and techniques from there too. In tandum, you should familiarize yourself with matplotlib, which is the plotting library it uses under the hood (and is by far the most popular plotting library.)

    For deep learning, I’d say PyTorch? Tensorflow used to be standard but its fallen out of favor compared to PyTorch. I don’t use either so I’m nit sure.








  • When I hopped on the home manager train I enabled starship (since it was just a couple lines to add) and I’m very happy with it. It has a couple small things out of the box that I really want. Mostly, its trimming my path so it doesnt take the full width of my terminal. I have it set so it only prints the lowest 3 directories and it wont print any directory higher than the current git repo Im in. IMO i hate all the little emojis but that was very easy to remove/disable. Its a very clean experience, and straightforward config (toml if not using nix).

    EDIT: here’s a pic of the path trimming. This is about as complicated as it gets. Also, I’m using catpuccin color scheme in urxvt.