Fix: Shorten embedding filenames using model_id_to_filename (#122)

[Rahul29999]

Description

Fixes Issue #122 by shortening embedding output filenames using a simplified function model_id_to_filename(), which extracts the final part of the HuggingFace model ID.

Changes Made

• Added model_id_to_filename() function to extract short model names
• Used it in output JSON file naming logic in cookbook/populate_embeddings.ipynb

Outcome

This resolves the issue of long file names like:
prompt_sentences-sentence-transformers-all-MiniLM-L6-v2.json
and converts them to shorter ones like:
prompt_sentences-all-minilm-l6-v2.json

Let me know if further refinements are needed!

[Rahul29999]

Hi @cassiasamp , I noticed the acknowledgment on the issue thread 👍. Just checking in here on the PR (#123) to see if it’s ready for merge or if you’d like me to make any further changes.

[cassiasamp]

Hi @Rahul29999, in the notebook I reviewed there was no change regarding the json filenames. Just to be sure, am I looking at the correct notebook?

There is also an specification about the issue, that I am probably going to fix because that might have been confusing.. but the issue mainly has to do with the generated and translated json files.

I believe there can also be an update to the populate_embeddings.ipynb cookbook, but the idea was to create a new notebook in which these json files names were shortened. Does that make sense?

[cassiasamp]

Hi @Rahul29999, I'm temporarily unassigning the #122 issue due to doubts regarding this PR. If there are any updates, I can reassign it later ✌️