tensorflow / tensorflow/tensorboard
it would be great to have a way to set seed in projector to get same UMAP/t-SNE visualization on every run
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 7.2k
- Forks
- 1.7k
- Avg merge
- 4d 22h
- Merged PRs (30d)
- 1
Description
Please describe the problem that you’re facing and the enhancements that
you’d like to see. Feel free to include screenshots or code samples.
Currently, tensorboard produces different embedding projections on every run. But it would be cool to pass some seed while running tensorboard --logdir ... to make tensorboard produce the same graphs for UMAP & t-SNE on every run.
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
Start from the TensorBoard entry point invoked with tensorboard --logdir ... and trace how UMAP and t-SNE projections are configured. Define how a user-provided seed should reach both projections, then verify that repeated runs produce the same embedding graphs.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- data-visualization
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100