Fr0zencr4nE/Glyph
Glyph scaling transfer — exact-image deduplication Upload in progress for scaling training on another cluster. Current release Input:1,713,825 historical native-quality AnyText candidate records. Output:1,459,580 unique images;254,245 duplicate image copies removed. Grouping:SHA256 of complete decoded512×512 RGB images. No center-crop/pHash/near-duplicate grouping. Per latest user instruction,no additional test-overlap or private-heldout exclusion is applied.… See the full description on the dataset page: https://huggingface.co/datasets/Fr0zencr4nE/Glyph.
| Author / organization | Fr0zencr4nE |
|---|---|
| Downloads | 949 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 23 October 2026 |
| Last updated at source | 24 October 2026 |
Popularity, last 3 days
949 now (+949 since 2026-10-23). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 23 Oct 2026 Description: Glyph scaling transfer — upload in progress This repository transfers prepared scene-text data to a new training cluster. It does not contain model weights. AnyText candidate pool 1,713,825 historical native-quality candidates: English grou → Glyph scaling transfer — exact-image deduplication Upload in progress for scaling training on another cluster. Current release Input:1,713,825 historical native-quality AnyText candidate records. Output:1,459,580 unique images;254,245 dupli
- 23 Oct 2026 Description: none → Glyph scaling transfer — upload in progress This repository transfers prepared scene-text data to a new training cluster. It does not contain model weights. AnyText candidate pool 1,713,825 historical native-quality candidates: English grou
Changes detected by Silytics between refreshes of the source.
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 25 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.