NVIDIA-Merlin / NVIDIA-Merlin/Merlin
How to re-train the already trained model with new small dataset.
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 907
- Forks
- 129
- PR merge metrics
- No merged PRs in 30d
Description
❓ Questions & Help
Details
I trained DeepFm model for recommendation with all of our data, now after few days have passed, I don't want to do a full re-train of the model with old + new data combined rather perform frequent updates with only the new data on an everyday basis how do we do transfer learning sort of thing here?
Model used - DeepFM for ranking.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names only the DeepFM ranking model and provides no files or tests. Start by checking the project's DeepFM training documentation and examples for supported incremental or transfer-learning workflows. Done means documenting whether daily updates using only new data are supported and the exact usage, or clearly stating that full retraining is required.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 15/100