From the course: Introduction to Transformer Models for NLP
Join today to access over 24,800 courses taught by industry experts.
Other LLMs and semantic search with OpenAI embeddings
From the course: Introduction to Transformer Models for NLP
Other LLMs and semantic search with OpenAI embeddings
- Let's turn our attention away from using Playgrounds to talk to ChatGPT and GPT-3.5 to another service provided by OpenAI called Embeddings. We've talked about Embeddings before! It's like, for example, when we used BERT to vectorize text into a vector to create a retrieval system or a semantic search system. If we look at the documentation for OpenAI Embeddings, they say, they talk about all the different use cases you can use text vectors for, including semantic search! And clustering, recommendations, anomaly detections, the list goes on! Ending with a tried and true NLP task classification. So, the ability to turn text into vectors as we've seen before, is really what opens up all of these tasks to anybody. So, revisiting our previous example of semantic search, we've seen this image before where we took a BERT model, in that case, a Siamese BERT model, we used it to create a database full of context, in that case, it was a textbook about insects. We used a BERT model to match…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.