Skip to content

docs: clarify Hugging Face loading for Kronos models#271

Open
chethanuk wants to merge 1 commit intoshiyu-coder:masterfrom
chethanuk:docs/issue-240-hf-loading
Open

docs: clarify Hugging Face loading for Kronos models#271
chethanuk wants to merge 1 commit intoshiyu-coder:masterfrom
chethanuk:docs/issue-240-hf-loading

Conversation

@chethanuk
Copy link
Copy Markdown

What changed

  • pin the README Hugging Face loading example to known tokenizer/model revisions
  • document the supported loading path via the local model classes and note force_download=True for stale cache mismatches

Why it matters

  • gives users a repo-supported loading snippet for the released Hub artifacts
  • addresses confusion around config.json and initialization errors when local caches are out of date

Validation

  • git diff --check
  • confirmed config.json exists in NeoQuasar/Kronos-Tokenizer-base and NeoQuasar/Kronos-small
  • confirmed the pinned revisions exist on the Hugging Face Hub

Fixes #240

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Question] How to load pre-trained models from HuggingFace? Missing config.json

1 participant