Text Classification
Transformers
English
Japanese
hallucination-detection
groundedness
rag
guardrails
rule-based
not-a-neural-model
Instructions to use NagaYu/claimcheck-rules with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NagaYu/claimcheck-rules with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="NagaYu/claimcheck-rules")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("NagaYu/claimcheck-rules", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| # ClaimCheck - deliberately minimal. | |
| # | |
| # All verification is local, deterministic, pure-Python. There is NO torch, | |
| # transformers, sklearn or spacy here, and there must not be: this runs on a | |
| # Hugging Face Spaces free CPU Basic box (2 vCPU / 16GB / ephemeral disk) and | |
| # the whole design depends on staying inside that budget. | |
| # | |
| # huggingface_hub is used ONLY for the optional relevance enrichment in | |
| # section (L) of app.py. If you never set HF_TOKEN it is never called, and the | |
| # gate is fully functional without it. | |
| gradio==6.22.0 | |
| huggingface_hub>=0.34,<2 | |
| pandas>=2.0,<3 | |