Great job following the course up to here! To recap, in this chapter you:
- Learned the basic building blocks of a Transformer model.
- Learned what makes up a tokenization pipeline.
- Saw how to use a Transformer model in practice.
- Learned how to leverage a tokenizer to convert text to tensors that are understandable by the model.
- Set up a tokenizer and a model together to get from text to predictions.
- Learned the limitations of input IDs, and learned about attention masks.
- Played around with versatile and configurable tokenizer methods.
From now on, you should be able to freely navigate the 🤗 Transformers docs: the vocabulary will sound familiar, and you’ve already seen the methods that you’ll use the majority of the time.