Purpose: ml/ is the canonical place for learning material, tutorials, implementations, and lightweight projects. It should contain both polished docs (for readers) and raw notes (for personal reference) in separate subfolders.
Quick index (create or verify these exist):
ml/01-docs/docs/— Polished tutorials, course-like writeups intended for readers.ml/03-notes/notes/— Raw personal notes, book notes, exploratory writeups.ml/02-implementations/implementations/— Code and notebooks with algorithm implementations.ml/04-kaggle/kaggle/— Competition exercises.ml/05-projects/projects/— End-to-end projects and pipelines.ml/06-scripts/scripts/andml/07-utils/utils/— Small reusable code modules.
How I suggest you use this folder:
- When experimenting, put drafts in
ml/03-notes/notes/. - When something becomes a tutorial or stable how-to, move the notebook to
ml/01-docs/docs/and update theml/01-docs/docs/README.mdwith a short description. - Keep datasets in
datasets/and reference them from notebooks using relative paths documented inml/README.md.
If you accept this organization I can prepare a small branch with recommended git mv commands for a subset of files you point out.