lyst / lyst/lightfm

Is there a way to save Dataset()?

Open
#698 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
5.1k
Forks
724
PR merge metrics
No merged PRs in 30d

Description

I have a trained model saved as a pickle file. My model has both user and item features.
When I load the model, I can make predictions for seen users by calling
pickled_model.predict(user_ids = [0], item_ids =[0])

But, I want to adjust it to a cold start problem, making predictions for new users. In this case, it requires to have dataset.fit_partial() , where the dataset should be fitted in the old dataset. It there a way to save and retrieve the dataset? Or we need to refit it every time before calling fit_partial()

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No file or test is named. Start by locating the Dataset class and its fit_partial workflow, then check how model serialization currently handles related state. Done means establishing and documenting a supported way to preserve the fitted dataset for cold-start predictions, or clearly recording that refitting is required.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.