Chatbot from the Command Line

It’s code breakfast time! This time we meet online, but with the positive vibe kept and some breakfast recipes shared! Here is what we will talk and code about.

**Chatbots from the Command Line**

In this code breakfast you get to build your first chatbot from the command line using Rasa open source and python. You’ll build an assistant that can;

  • detect what the intent of the user is
  • parse the text to retreive important references
  • use custom python code in the conversation
  • stick to predefined stories (we’ll see a huggingface demo to demonstrate why this is a nice thing)
  • actively learn from your conversations

We’ll also briefly discuss topics in NLP in general, like named entity recognition but the main goal is to get a working assistant running locally. Rasa works on top of tensorflow 2 but you won’t need to be familiar with it to get it to run.

If you’re new to python, or if you want to make a head start, then you may appreciate these two installation guides for Rasa;
– macOS: https://www.youtube.com/watch?v=fqzsE70Rvr0
– windows: https://www.youtube.com/watch?v=4ewIABo0OkU

About the trainer:
Vincent D. Warmerdam is a Research Advocate at Rasa where he helps the research team explain and explore modern nlp algorithms for digital assistants. You may have seen him before at PyData. You can usually find him on stage defending common sense over hype but you may also have seen some of his open source projects (scikit-lego, evol and whatlies), some of his online videos (spaCy, algorithm whiteboard) and he also runs a few projects on the side (calmcode.io, thismonth.rocks, drawdata.xyz, dearme.email) as well as a blog over at koaning.io.

Group: PyData Amsterdam
Friday 1 mei 2020
9:00 tot 10:30 CEST

April 23, 2020