We are a friendly and enthusiastic group of Central Ohio Pythonistas. We meet most Thursday evenings for informal dojos, occasionally for lunch on various days of the week, and have a monthly meeting on the last Monday of the month. To have lunch at your favorite place, announce the time and place on the mailing list.
After our monthly meetings we like to adjourn to Brazenhead to enjoy good food, good drinks, and good conversation.
We have code on our github repository, especially of challenges where people solved problems using Python in very different ways.
Dojos are informal Python group study sessions. Complete beginners, experts, and everyone else are welcome to the Dojos. Bring your Python questions and problems. Digressions from Python are common.
Please bring outlet strips and extension cords.
June 26, 2017, 6:00 p.m.
20:00 Brazenhead Irish Pub
Come and learn, share, grow, meet new people, and visit old friends at our monthly meeting! We'll be talking about the Python programming language and anything that intersects it, and the cool stuff you can do with it.
This month, Jeff Klukas will be presenting "Machine Learning in Production with Scikit-learn".
We'll be discussing Simple's implementation of a Python microservice for classifying incoming chat messages by subject category, enabling our customer relations agents to develop specializations and onboard more quickly.
We'll walk through a bit of the code for our model and what the interface looks like in scikit-learn for training a model, persisting it to disk, and requesting a prediction.
Once we understand the shape of interacting with scikit-learn, we'll take a look at wrapping it in a Flask app and the concerns about understanding how that application is behaving in production. This includes performance metrics, logging results to a database, and degrading gracefully when things go wrong.
We'll then switch gears to talk about all the work that needs to happen outside of the application itself. We use a separate framework to execute scheduled jobs, periodically retraining the model on new records in our data warehouse or testing out a new iteration of the model code. We evaluate the performance of the models based on historical data, and then update the model running in production when we find a better-performing configuration.
Finally, we'll discuss how other companies approach the problem of serving predictive models in production. Varying concerns around performance needs, security constraints, and technical expertise can vastly change the shape of the solution.
Please RSVP so our generous host knows how much pizza and beer to order.