1. 5
  1.  

    1. 6

      Fine is neat, but I’d rather gargle buckshot than let the clowncar of Python ML stuff anywhere near a clean, self-respecting Elixir deployment. I sure as shit don’t want it sharing the same address space as my BEAM VM.

      I’m glad they’re trying stuff though.

      1. 2

        I think we all understand the implicit subtext of “this is fine”.

        Does the page load for you? I just get an ERR_SSL_PROTOCOL_ERROR from dashbit.co.

        1. 1

          Yeah I mean I love what they did with Nx/Bumblebee but there was no way they’re going to keep up with the Python ML ecosystem. For example, a year ago I wanted the YOLO model but it wasn’t available at the time. So I also appreciate that they give us an out with this.