Through the publication of the DocLayNet dataset (IBM Research) and the publication of Document Understanding models on Hugging Face (for example: LayoutLM series and LiLT), a Document Understanding model at line level had already been published: a LiLT base model finetuned on the dataset DocLayNet base with overlap chunks of…

Through the publication of the DocLayNet dataset (IBM Research) and the publication of Document Understanding models on Hugging Face (for example: LayoutLM series and LiLT), a Document Understanding model at paragraph level has been fine-tuned and published: a LiLT base model with overlap chunks of 512 tokens that uses the…

Through the publication of the DocLayNet dataset (IBM Research) and the publication of Document Understanding models on Hugging Face (for example: LayoutLM series and LiLT), a Document Understanding model has been fine-tuned and published: a LiLT base model at line level with overlap chunks of 384 tokens that uses the…

The publication of the DocLayNet dataset (IBM Research) and that of Document Understanding models on Hugging Face (for example: LayoutLM series and LiLT) allow (at last!) the training of such models on all documents with text (for example: PDFs, slides, images with text) with labels that interest the greatest number…

After creating different formats (small, basic, and large) to download DocLayNet (formats small/base/large that also help using DocLayNet data in Hugging Face notebooks on finetuning document layout models), it was important to have an APP for viewing annotated images with labeled bounding boxes of paragraphs and lines. Indeed, this visualization…

Document AI is still a new area of NLP but affects all businesses and individuals. It consists of using AI models to visually and textually understand the content of documents such as PDFs. Thus, it is then possible to categorize the text, images and tables of documents (header, main text…

Paga por um serviço online para obter transcrições de texto de seus arquivos de áudio? E porque não usar um modelo Whisper da OpenAI para fazer esse trabalho… de graça! Precisa especializar um modelo Whisper para as peculiaridades linguísticas de seus arquivos de áudio? Não tem problema, existem scripts e…

Como pode obter inferências mais rapidamente (e de uma forma muito simples!) com seus modelos Transformer existentes (ou seja, sem precisar treiná-los novamente) e, assim, reduzir os custos do servidor de inferência? A solução vem através do código: BetterTransformer da Hugging Face! …

Este artigo explica por que e como usar uma função de perda ponderada para treinar um modelo de classificação de texto. Notebook: Text_Classification_on_GLUE_with_weighted_Loss.ipynb Uma das tarefas mais comuns em NLP Conscientemente ou não, uma das atividades intelectuais mais frequentes nos negócios é a classificação de textos. …