Datasets:
|
Download quality_tools/README.md from SEE7/AstheralDataset: direct link, hf CLI and curl.
- Browser
- Download file 536 Bytes
-
https://huggingface.co/datasets/SEE7/AstheralDataset/resolve/main/quality_tools/README.md
- Command line
-
hf download hf://datasets/SEE7/AstheralDataset/quality_tools/README.md
-
curl -L -o README.md https://huggingface.co/datasets/SEE7/AstheralDataset/resolve/main/quality_tools/README.md
536 Bytes
| # Reproduce the dataset repair | |
| Run from an empty working directory on a CPU machine: | |
| ```sh | |
| pip install -r quality_tools/requirements.txt | |
| python quality_tools/audit_public_dataset.py | |
| python quality_tools/check_generated_code.py | |
| python quality_tools/score_alignment_cpu.py | |
| python quality_tools/finalize_repair.py | |
| ``` | |
| The source dataset and embedding model are pinned to the revisions in the quality report. Outputs are written to `dataset-repair/publish`. These scripts read public resources and do not require an access token or GPU. | |