tensorflow / tensorflow/tensorboard
See model scoring of a few Examples
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 7.2k
- Forks
- 1.7k
- Avg merge
- 4d 22h
- Merged PRs (30d)
- 1
Description
A really useful feature would be to be able to select a random sample of ~100 Examples and see how the model scores them. It could be over time, or just at the current moment.
This would be beneficial for analyzing what the model cares about, and what kinds of Examples it does well or poorly on.
Even better, maybe you could be able to tweak an Example and see how it affects the model's score.
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 by locating the existing model-scoring and Example workflows in TensorBoard. Determine how a random sample of about 100 Examples could be selected and displayed, whether scores can be compared over time, and how an edited Example would be evaluated. Done should include a defined interface for inspecting scores and a clear way to assess the requested analyses.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- analytics, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100