lyst / lyst/lightfm

Item similarity without using `interactions` data?

Open
#535 0 comments 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

Not an issue, just a question. In the `fit` method I see that `user_features` and `item_features` are optional.

```
fit(interactions, user_features=None, item_features=None, sample_weight=None, epochs=1, num_threads=1, verbose=False)
```

Wondering for a totally cold start problem where all I have are `item_features` if it is possible to fit a model, e.g. something like:

```
fit(item_features, interactions=None, user_features=None...)
```

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

Start with the fit method signature and its handling of interactions, user_features, and item_features. Determine whether fitting from item_features alone is supported; done should be a documented answer or a clearly scoped implementation plan for cold-start training.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.