Currently, Scikit-tt can't be run as-is after pulling the repo, since it uses np.infty, which was removed in numpy 2.0.
At least on my machine the requirements get evaluated to use numpy>=2.0.
Solution
Fixing the numpy version is very easy to do manually but I figure it might be nicer to have it in the requirements and done automatically.
Situation
Currently, Scikit-tt can't be run as-is after pulling the repo, since it uses np.infty, which was removed in numpy 2.0. At least on my machine the requirements get evaluated to use numpy>=2.0.
Solution
Fixing the numpy version is very easy to do manually but I figure it might be nicer to have it in the requirements and done automatically.