We provide several tutorials to help users get familiar with Bootleg.
In this tutorial, learn how to use Bootleg for end-to-end inference. We start from text data and show how to detect mentions and then link them to entities. We also show how to use Bootleg for "on-the-fly" disambiguation of individual sentences.
In this tutorial, we show how to use Bootleg for "on-the-fly" disambiguation of individual sentences.
In this tutorial, learn how to modify and change the entity database associated with a Bootleg model. We start from the downloaded entity profile data and show how to add/remove entities and change type and relation mappings. We then show how to fit an existing model to this profile and load it into a new annotator (or use it on your own data!).
In this tutorial, we will introduce you to how to take a pretrained Bootleg model and generate entity representations. The next tutorial shows you how to use them in a downstream model.
In this tutorial, we show you how to integrate Bootleg embeddings into a downstream LSTM model and SPAN-BERT model.
In this tutorial, learn how to train a Bootleg model on a small dataset. This will cover input data formatting, data preprocessing, and training.
In this tutorial, learn how to use distributed training to train a Bootleg model on the full English Wikipedia save (over 50 million sentences!). You will need access to GPUs to train this model.