Historical score exporter: STATIC REFERENCE ONLY

WARNING: DO NOT RUN OR IMPORT THE ARCHIVED PROGRAM OR ITS TESTS. The legacy program includes opt-in upload and a private-quota fallback branch that calls update_repo_settings(private=False). That historical behavior is NOT authorized by this archive and conflicts with the current preservation rule against changing existing private repository visibility. The archived tests also encode old fallback behavior; they are evidence of history, NOT approved operating policy.

Exact Git source/test bytes are preserved as .py.txt text references, not installed as a runnable package. This publication performed AST/content review only; it did not execute, import or run tests from either file. No --upload, fallback, repository visibility change or model export was invoked. Use current independently reviewed publication tools instead, with explicit authorization, locks, immutable receipts and rights checks.

The reference illustrates validation guards, model/processor allowlists, configuration sanitization and old model-card construction. It includes generic privacy-regex tokens and synthetic private-path fixtures; these are not actual private machine paths. No weight, optimizer, dataset, sample-level feedback, credential or original run log is included. STATIC_REVIEW.json pins exact source hashes, provenance and the unsafe historical branch. Read LICENSE and NOTICE. UPLOAD_MANIFEST.json covers every other file.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support