Please register with this Google Form https://goo.gl/zsPNvd and we will get back with a confirmation. Just Eventbrite RSVP is not enough.

Machine learning has become critical not just in developing a new application but also in improving the quality of business decisions. We have many tools and frameworks that allow us to experiment with data and build machine learning models, but productizing these models still appears to be a barrier. Data scientist need a better way to deploy their work in production and application developers need a standard way of accessing these services.

In this workshop, you will learn how to build a small machine learning model end-to-end and deploy it in production. We will try and cover a small example of deep learning if we have some time.

Date: October 14, 2017, Saturday

Time: 11 AM

City: Hyderabad

Venue: coming up shortly

The workshop is FREE, but it will be invite based due to limited seating. See link above

Outline

Section I - Overview

• Tour of the Python data science ecosystem

Section II - Data Ingest & Wrangling

• Data ingestion
• Data munging using Pandas (Hands-on)
• Creating data pipelines using workflow (Hands-on)

Section III - Building Machine Learning Model

• Build a machine learning model using Scikit: Estimating House Prices Using (Hands-on)
• Build a deep learning model with Keras: Tagging images using deep networks (hands-on, more an introduction)
• Training model in cloud

Section V - Models in Production

• Machine learning models in production
• Deploying your models into production (Hands-on)
• Creating machine learning pipeline using workflow (Hands-on)

Open Discussions

About the workshop

What will we cover?

• Life-cycle of a machine learning application - right from data munging to deploying on cloud
• Build machine learning models using Scikit-Learn and Keras
• Issues in building an end-to-end pipeline and productionizing, how to overcome them
• Build and deploy end-to-end models to production on public cloud

What will we NOT cover?

• Theory or mathematics of machine learning, we can point to external resources for reference
• Discussion on algorithms or comparison of performance
• Non-Python machine learning ecosystems such as R, Scala
• Big data technologies such as Hadoop

Who is this workshop for?

• A machine learning practitioner
• Developer exploring machine learning
• We welcome startups as well as data science teams at enterprises

Some prior experience will be useful. Focus of this workshop is “Putting it all together” rather than learn basics of machine learning models

Workshop will be invite based, please express your interest here https://goo.gl/zsPNvd

Pre-requisites

• Participants must be comfortable with Python programming
• Familiarity with Python Data Stack is ideal, should be able to run Jupyter Notebook
• Basic understanding of algorithm

Software

Please install Anaconda for Python 3

Material

Links will be uploaded here soon.