Patrick Haller

By checking the input of the radio box (for male(1) or female(2)), the conversion from string to int is missing so it is giving back an error message( invalid operation:...

Print Debugging

3

Almost everything implement but the interface ```python from frosch import fprint as fp fp(some_var) ```

Automatic injection of frosch in python interpreter

1

Either write a script, that hooks itself into the python interpreter so frosch is always executed or find another way to its auto hooking by simply pip installing frosch

More configurability

2

This means for colours and maybe symbols. Frosch will in the future allow for way more configuration options for different purposes. We should therefore come up with a better config...

Replace print with logging in model training

2

**Is your feature/enhancement request related to a problem? Please describe.** When using the ModelTrainer, everything is printed out through a logger, except for when the learning rate is reduced. This...

PromptNode not usable if AutoTokenizer fails

1

**Describe the bug** Using the PromptNode with the HF Inference API Endpoint with the model `timdettmers/guanaco-33b-merged` will throw a recursion error, due to a (at least) missing `unk_token` of the...

Important for skip list is the memory usage (compared to linked lists etc.), especially depending on the max layers. Some more infos of it in runtime would be nice Would...

Add support for other Python LSP server

1

Python LSP server: * [x] pyright * [x] jedi * [x] pylsp * [ ] [python-lsp-server](https://github.com/python-lsp/python-lsp-server) * [ ] [ruff-lsp](https://github.com/charliermarsh/ruff-lsp)

Not happy with just using raw tables with strategies and paths. Maybe use metaclasses venvs and be more descriptive and less hacky