google-gemini / google-gemini/gemini-cli
Triage tracker: 5 oldest unlabeled open issues in huggingface/datasets
- Dominant language
- TypeScript
- Stars
- 107k
- Forks
- 14.6k
- Avg merge
- 2d 3h
- Merged PRs (30d)
- 45
Description
## Summary
Triage workboard for the **5 oldest open, unlabeled issues** in the upstream [huggingface/datasets](https://github.com/huggingface/datasets) issue tracker. None of these issues has any label attached, which means they have never been triaged. This issue collects them in one place so the team can work through them as a board: apply labels upstream, assess priority, and check whether each one is still reproducible or relevant.
**Source:** snapshot of the huggingface/datasets issue tracker (`github-issues-corpus.jsonl`: 3,019 records → 984 issues after excluding pull requests → 325 open → 120 open with zero labels). The 5 entries below are the oldest of those 120, sorted by creation date (ascending).
## Issues needing triage (oldest first)
| # | Issue | Title | Created (UTC) |
|---|-------|-------|---------------|
| 1 | [huggingface/datasets#373](https://github.com/huggingface/datasets/issues/373) | Segmentation fault when loading local JSON dataset as of `#372` | 2020-07-10T15:04:25Z |
| 2 | [huggingface/datasets#375](https://github.com/huggingface/datasets/issues/375) | TypeError when computing bertscore | 2020-07-10T20:37:44Z |
| 3 | [huggingface/datasets#376](https://github.com/huggingface/datasets/issues/376) | to_pandas conversion doesn't always work | 2020-07-10T21:33:31Z |
| 4 | [huggingface/datasets#413](https://github.com/huggingface/datasets/issues/413) | Is there a way to download only NQ dev? | 2020-07-18T10:28:23Z |
| 5 | [huggingface/datasets#433](https://github.com/huggingface/datasets/issues/433) | How to reuse functionality of a (generic) dataset? | 2020-07-24T17:27:37Z |
## Workboard checklist
- [ ] [#373 — Segmentation fault when loading local JSON dataset as of `#372`](https://github.com/huggingface/datasets/issues/373) — created 2020-07-10T15:04:25Z
- [ ] [#375 — TypeError when computing bertscore](https://github.com/huggingface/datasets/issues/375) — created 2020-07-10T20:37:44Z
- [ ] [#376 — to_pandas conversion doesn't always work](https://github.com/huggingface/datasets/issues/376) — created 2020-07-10T21:33:31Z
- [ ] [#413 — Is there a way to download only NQ dev?](https://github.com/huggingface/datasets/issues/413) — created 2020-07-18T10:28:23Z
- [ ] [#433 — How to reuse functionality of a (generic) dataset?](https://github.com/huggingface/datasets/issues/433) — created 2020-07-24T17:27:37Z
## Notes
- These are **upstream** issues living in `huggingface/datasets`, not in this repository. Any triage action (labeling, prioritizing, closing) must be carried out in the upstream tracker; this issue only serves as the tracking workboard.
- Data was taken from a static snapshot, so the current upstream state (labels/status) may have changed since capture. Verify each issue upstream before acting.
Contributor guide
Research direction
Start with the issue body and github-issues-corpus.jsonl, then verify the five linked issues in the upstream huggingface/datasets tracker. Review each issue's current labels, priority, reproducibility, and relevance, and update the upstream tracker as appropriate. The work is done when all five checklist entries have been assessed and the upstream state is reflected.
Written by the indexing model from the issue text.
Assessment
- Domain
- tooling
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100