Instructions to use Sennodipoi/LayoutLMv3-kleisterNDA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Sennodipoi/LayoutLMv3-kleisterNDA with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="Sennodipoi/LayoutLMv3-kleisterNDA")# Load model directly from transformers import AutoProcessor, AutoModelForTokenClassification processor = AutoProcessor.from_pretrained("Sennodipoi/LayoutLMv3-kleisterNDA") model = AutoModelForTokenClassification.from_pretrained("Sennodipoi/LayoutLMv3-kleisterNDA") - Notebooks
- Google Colab
- Kaggle
Add Kleister NDA performance metrics
#2
by mbuet2ner - opened
Hi Malte, glad to know that people are interested in this :)
You're definitely right, I should put more info in the model cards. I'm planning to update them all after my thesis defense.
In the meanwhile, if you are interested in LayoutLMv3 for Kleister-NDA, I was able to reach a f1 score of 0,94 using the same segmentation strategy used by the original authors for FUNSD i.e. using labels to create segments. However, in the case of Kleister-NDA I believe that this strategy significantly helps the model and thus it severely impacts the evaluation of the model.
Hi, I added the resultsfor all the LayoutLM models.
Cheers,
Alessandro
Sennodipoi changed discussion status to closed