10

NLP 2.0: Using Transformers to Generate Text

As we saw in the previous chapter, the NLP domain has seen some remarkable leaps in the way we understand, represent, and process textual data. From handling long-range dependencies/sequences using LSTMs and GRUs to building dense vector representations using word2vec and friends, the field in general has seen drastic improvements. With word embeddings becoming almost the de facto representation method and LSTMs as the workhorse for NLP tasks, we were hitting some roadblocks in terms of further enhancement. This setup of using embeddings with LSTM made the best use of encoder-decoder (and related architectures) style models.

We saw briefly in the previous chapter how certain improvements were achieved ...

Get Generative AI with Python and TensorFlow 2 now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.