Instructions to use Faradaylab/aria-doc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Faradaylab/aria-doc with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "question-answering" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("question-answering", model="Faradaylab/aria-doc")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForQuestionAnswering tokenizer = AutoTokenizer.from_pretrained("Faradaylab/aria-doc") model = AutoModelForQuestionAnswering.from_pretrained("Faradaylab/aria-doc", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from Faradaylab/aria-doc: direct link, hf CLI and curl.
- Browser
- Download file 823 Bytes
-
https://huggingface.co/Faradaylab/aria-doc/resolve/main/README.md
- Command line
-
hf download hf://Faradaylab/aria-doc/README.md
-
curl -L -H "Authorization: Bearer $HF_TOKEN" -o README.md https://huggingface.co/Faradaylab/aria-doc/resolve/main/README.md
823 Bytes
metadata
license: gpl
datasets:
- etalab-ia/piaf
language:
- fr
- en
metrics:
- accuracy
pipeline_tag: question-answering
tags:
- biology
- legal
- French
- France
Aria Doc is a model we created for Documents Q/A. It has been trained on a high quality dataset with over 3000 rows of context,question and answers from Etalab. You can find the dataset used for training here https://huggingface.co/datasets/etalab-ia/piaf . Aria Doc has a great performance in French related Q/A and data extraction in french language.
- Aria Doc est un modèle que nous avons créé pour l'extraction de données sur des documents et les besoins de questions/réponses sur des données spécifiques. Le modèle a été entraîné sur plus de 3000 exemples fournis par le Dataset PIAF. Les performances d'Aria Doc sont optimales en Français.