What we're about

If you are a data scientist or have an interest in data science this is the meetup group for you!

The goal of this meet up is to prepare attendees for doing data science as well as expand the set of skills of practicing data scientists. The meetup will alternate every week between lecture style and hands on style.

All lectures will be recorded and uploaded to youtube. All hands on style nights will be linked to on github.

The topics covered will include:

* Deep Learning
* Classical Statistics
* Tree Based Learning
* Bayesian Statistics
* Bayesian Learning
* Data Science at Scale
* Engineering for Data Science
* Ethical Data Science
* Data Cleaning and Preparation
* Soft Skills for Data Scientists

All lectures will be in a combination of R, Python, Julia, Scala, Javascript and Stan.

Upcoming events (1)

How to Build and Operationalize ML Pipeline

11 Times Square

In this session, Eric (Data Scientist at Microsoft, https://www.linkedin.com/in/ericschles) and Razi (Senior Software Engineer at Microsoft https://www.linkedin.com/in/razirais) provide a walkthrough about how to build and operationalize an ML pipeline for text classification. The idea is to help attendees with little or no ML background to understand moving parts of a typical ML project. Please do note that this is a technical session and most useful for people with hands-on software engineering experience. Data science experience is not needed but basic knowledge of it won't hurt. Also, we will be building the ML pipeline from scratch to provide a full understanding of what goes inside building and running a ML pipeline. For these reasons, we wont be covering Kubeflow, MLflow etc in this session. These are all great tools but you may learn more by doing it yourself initially and then go with a tool of your liking later. Topics covered in the sessions: 1. An Introduction to Supervised Learning - We will start the session with a basic introduction to supervised learning. 2. Identify labels and features - We will focus on how to use active learning to manually label examples in lieu of a labeled data set. 3. Creating models - Here we will show a demo of how to make use of sklearn to build a model, we will briefly go over the API and how to assess model quality 4. Generating predictions - Here we will demonstrate how to use the sklearn model in production to generate new labeled data, from the algorithm 5. Packaging Models - We will be using containers to package the models and other artifacts that eventually run as part of the ML pipeline (next item) 6. Building ML pipeline using Azure Logic Apps (Workflow) - Demonstrate how to build ML pipeline using Azure Logic App and ACI (Azure Container Instances). Demo also cover automation of the pipeline.

Photos (2)