Michael Dales

Thanks for the feedback @andreynering. I've not forgotten about this, but I will say that it is harder that I initially (naively :) assumed. My deployment is with Postgres, and...

I've had a little look here at the options others have used, and what I've ended up doing in the product I'm working on is use an advisory lock. This...

FWIW, the testing in my application for the lock wasn't too bad. I was worried I'd need to augment gormigrate to let me synchronise things properly to ensure testing clashing...

Just to be clear, I don't expect winget to install things inside WSL, but WSL is my default shell, and I run other windows commands from inside there. I'd still...

I'd have to investigate using subtrees for Claudius because submodules are not supported/working. However, I find this a suboptimal solution for my particular use case, and would much rather see...

Just some notes on this incase it saves someone else so pain. You can work around this by switching to the home-brew llvm, but if your ultimate aim is to...

Locally trying to repro, within the same Python virtual environment I can just downgrade MLX only to the previous release and things work, but then upgrading MLX only causes the...

I changed my CI job to use `mlx[cpu]` and that no longer has the issue. However, using just `mlx` on macOS I do still have the issue. Successful run is...

From my failing setup: ``` $ pip show mlx-metal Name: mlx-metal Version: 0.26.5 Summary: A framework for machine learning on Apple silicon. Home-page: https://github.com/ml-explore/mlx Author: MLX Contributors Author-email: [email protected] License:...

On a fresh virtual environment everything is working :/ ``` $  python3 -m venv /tmp/venv lattebit  ~  $  . /tmp/venv/bin/activate.fish (venv) venv  lattebit  ~...