# CRlibm test vectors: asin, atan, cosh and more, correctly rounded in four rounding modes (`crlibm-testdata`) Kind: material set (task-ready). Units admitted from one source for building tasks, counted in units: each unit is a group of files its upstream publishes as one thing (`unit_id`). 11 units (13 files) from 1 publisher. Summary from sample files: not judged yet. The text above gives this release's counts of the source. ## Domain and form - Domain: `D4.3` (GPU kernels and operators). - Form: `test-suites`. Test cases shipped with software, specifications and tools (conformance suites, regression tests, test corpora, benchmark programs), each with its expected result. ## Unit and scale - 11 units (13 files). - Rows: 13. - Size: Recorded size 1.82 MB; deduplicated size 1.82 MB; size check: 100.0% match (13 of the 13 rows read compared). - Fetch check: all 13 sampled rows read fetched and matched their recorded size and hash (read 2026-11-14). - Batches: 1 (`2026-10-ml-serving-configs-2`). Collection: complete. ## File types and sizes | Type | Rows | Size | |---|---:|---:| | `testdata` | 11 | 1.80 MB | | `(none)` | 1 | 230 B | | `md` | 1 | 21.6 KB | Size quantiles: unknown (no size distribution recorded for this source). ## License and access - License: `GPL-2.0-only`; license tier `share-alike` (reason code `copyleft`); training use: yes; redistribution: conditional 85%, yes 15%; files may be shipped: conditional. - License statement: The units are under GPL-2.0-only, as recorded from their publishers; the licence file is campaign_files/2026-10-ml-serving-configs-2/crlibm-testdata/LICENSE. - Login: no. - How to fetch: not recorded on the card; the row's uri is fetched as it is. ## Freshness - Rows are counted for the release of 2026-11-20; the card's own sample and counts date from the card's writing (it has no snapshot date of its own). ## Readiness - Readiness level: not judged yet. - Card review: `draft` (a `draft` card has not been reviewed as a whole). - Personal information read from samples: not judged yet. ## Query this source at a fixed release Pinned to the release tag `release-2026-11-20.1`. ```sql SET VARIABLE files = (SELECT list('hf://datasets/Battam/training-data-foundation@release-2026-11-20.1/' || f) FROM (SELECT unnest(data_files) AS f FROM 'hf://datasets/Battam/training-data-foundation@release-2026-11-20.1/meta/catalog.parquet' WHERE source = 'crlibm-testdata')); SELECT uri, ext, size_bytes, lower(hex(sha256)) AS sha256 FROM read_parquet(getvariable('files')) WHERE source = 'crlibm-testdata' LIMIT 10; ``` ## Sample files - `campaign_files/2026-10-ml-serving-configs-2/crlibm-testdata/README.md` - `campaign_files/2026-10-ml-serving-configs-2/crlibm-testdata/LICENSE` ## Parts not to use for tasks - Parts to leave out: not judged yet. - Benchmark overlap `none`: every data file is compared with the benchmark list, and the default profile of `tools/fetch.py` removes none of the rows. The card [`crlibm-testdata.yaml`](crlibm-testdata.yaml) (card_version 5) holds every field with its evidence; values shown here as not judged yet have no judgment on the card so far.