This starter notebook is provided by the Keras team.
In this competition, our aim is to develop an AI model that can score student essays. This competition is actually an updated version of an old one that took place over a decade ago. In this version, we aim to improve upon essay scoring algorithms to enhance student learning outcomes. This notebook will guide you through the process of fine-tuning the DebertaV3 model using Ordinal Regression/Classification to score student essays using KerasNLP.
Did you know: This notebook is backend-agnostic, which means it supports TensorFlow, PyTorch, and JAX backends. However, the best performance can be achieved with JAX. KerasNLP and Keras enable the choice of the preferred backend. Explore further details on Keras.
Note: For a deeper understanding of KerasNLP, refer to the KerasNLP guides.
Training and Inference notebook is available in /notebooks and you can also access it on Kaggle here.