As stated in other responses, Py to exe isn’t a reliable tool, and will likely put libraries into TCL for Windows, meaning that you would have to package it with an installer, which wouldn’t be so bad if it didn’t take up so much space. Packaging a 15GB blob of TCL just isn’t feasable.

And yes, Python is fast enough to do the job for a lot of things, but I disagree with the idea of ignoring deficiencies with a language because I like it. Though for the syntax, yes Python is fast, but in the scope of the languages around it, no it is objectively not very fast. Python is used for ML because it lacks the issues with R, has been around since the 90s, and has a close bond with C, not because it is fast. Maybe you should do some research into the two languages I compared Python with, Nim and Julia, and see what they’re about.