This will be a tutorial based talk with a demo at the end on how to use conversational language models to make an interactive chatbot. The models that we will use are Blenderbot from Facebook Research and distilbert from google research / huggingface. We will also use Transformers from Huggingface which is an easy to use packaging and API for the above systems and also go over using speech recognition and text to speech to make interaction more fluid. At the end there will be an interactive demo.
Learn how we built cross-application search at Tegus by integrating Tegus content into recently-acquired BamSEC's document search system, using AWS SQS, Elasticsearch, and Python.
An overview of property-based testing with hypothesis.
Contrasting the Python Programming language to Racket.
https://racket-lang.org
Craiyon.com is a front end for a neural network that takes in free-form text and generates images. It can be viewed as a "language" that "compiles" to produce a binary that is an image. This is one future of programming.
When does it make sense to use Rust? What are the top challanges a Python user will face when trying Rust for the first time?
In keeping with the proposed theme, I'd be happy to introduce R and compare/contrast it to Python. I'd like to talk especially about how well suited R is to Statistics, and how Pandas dataframes were inspired by R dataframes. I've used both in my career as a data analyst and now data engineer, and initially prefered R though I've come around to the Python side.
SQL is a declarative, recursive language. For this year's Ultimate Language Shootout, I'm going to talk about how SQL is the best language that's not Python.
Last time I talked about Micropython I didn't show any code.
This time I have code:
>>> from machine import Pin
>>> p=Pin("D13", Pin.OUT)
>>> p.value(1)
That turns on an LED. We will talk about what that means, how it happens and the implications of this is the software that makes electronics happen.
There are a few more lines of the same nature: check the state of a switch gives is the I in IO. For extra credit, we will ...
The best way to share machine learning models is through interactive web applications that let stakeholders try the models for themselves. By playing with the model through a graphical interface, stakeholders can develop an intuition for how the model works in a way that would be impossible by analyzing average performance metrics or hand-picked examples.
Building web applications for machine learning requires significant knowledge of web development (css, js) and web hosting which are not in the typical data scientist's tool box.
In this talk, we will introduce a gradio, an open source python library for building and sharing machine learning applications entirely in python.