OxML 2023 Financial Machine Learning sompetition notebooks with code.
Task 1: Given a set of pdf documents, Build an ESG document classifier that can take a document as an input, classify each page to be either E,S or G related
Task 2: Given a set of pdf as images, Build a table detector that can precisely locate the position of table from of a page document
Task 1: Finutune DistillBert on train/validation/test set
Task 2: Zero-shot DETR style table transformer.
- huggingface
- PyMuPDF