CA$49.50 – CA$229

[online] Premium Handson Workshop: Deep Learning- from Theory to Deployment

Event Information

Refund Policy

Refund Policy

Contact the organizer to request a refund.

Eventbrite's fee is nonrefundable.

Event description
Learn how to build and deploy deep learning models

About this Event

Workshop Overview

This 4-week workshop is designed to walk you through some of the concepts in deep learning as well as machine learning operations (MLOps) so that you can

  • design and train a deep learning model
  • deploy the model in a cloud-agnostic way
  • build an application on top of the deployed model (Capstone Competition)

Learning Journeys

  • I just want the content: you can learn at your own pace, just select the "content" when you check out
  • I'm new to Deep Learning: learn the basics, practice hands on, form a team, participate in the competition
  • I just want to participate in the capstone: if you have taken our previous deep learning workshop or otherwise know enough deep learning already, then select "capstone only" options when you check out

Capstone Competition

All participants have an opportunity to enter a competition by finishing and submitting all the requirements of the capstone project. The projects will be judged based on effort, readiness of the end product, performance of the product, and the practicality of the solution. The winner team members will receive a 100% refund, and all team members contributing to finished capstone projects will receive 10% refund.

Important Dates

Early Bird Deadline: April 9, 2020

Refund Deadline: April 12, 2020

Start Date: April 13, 2020

Last day to Join: April 19, 2020

Competition submission deadline: May 17, 2020

Time Commitment

  • Week 1: 3 hours individual work, 1 hour group activities, 0 hour capstone
  • Week 2: 3 hours individual work, 1 hour group activities, 1 hour capstone
  • Week 3: 3 hours individual work, 1 hour group activities, 1 hour capstone
  • Week 4: 0 hours individual work, 0 hour group activities, 1 hour capstone

Please note that this is the min requirement. You have to consider the amount of extra time you would need to keep up, or do group or individual work esp for the capstone

Please also note that all the group activities and capstone work will be done through video calls

Milestones

  • Week 1
    • go through the content
    • meet the participants
    • participate in group activities
    • submit your preferences about who you want to team up with
    • capstone teams finalized
  • Week 2
    • go through the content
    • participate in group activities
    • pitch your capstone project and get feedback
  • Week 3
    • go through the content
    • participate in group activities
    • demo the your capstone project Steel Thread
  • Week 4
    • go through the content
    • participate in group activities
    • demo your final capstone project
  • Week 5: final touches and submit for the competition (submission involves creating one 2-5 min videos of you explaining what you built, and demoing it)

Target Audience

Data Scientists, Machine Learning Engineers, Software Engineers

Prerequisites

  • Comfortable with pandas, sklearn, and familiar with PyTorch
  • Modelling environment: google colab
  • Theoretical knowledge assumed: It is expected that the participants are already familiar with the basic statistics, ML, neural network.

Learning Material

Participants will have access to the following learning material:

  • Slides
  • Hands on notebooks
  • Video recording of theory and code walkthrough
  • Access to slack for interactions with other participants and teaching staff
  • Live code walkthrough with TAs( addon)
  • [TBD based on demand] Live session with the instructors

Instructors

Please note that the instructors have provided the material but the sessions will be self-paced or TA led. There will be a session available with the instructors based the amount of demand. When you are checking out you can indicate your interest in a session with the instructor.

Amir Hajian

Director of Data Science at Scribd

Amir is an astrophysicist by training and a data scientist by trade. Amir is currently the Director of Data Science at Scribd. He worked as a Director of AI Research at Thomson Reuters before. Amir's job is to lead teams of scientists and engineers to design and build end to end products powered by machine learning, statistics, graph analysis and NLP. Before Thomson Reuters, Amir worked as a post-doctoral researcher at University of Toronto and Princeton University.

Brenden McGiven

Machine Learning Engineer at Chisel AI

Brendan is a software developer, turned machine learning engineer, passionate about both technology and utilizing machine learning to solve business problems. He has experience working with Python, PyTorch, scikit-learn, Spark and AWS. He has had the pleasure of working with companies from all over the world (Canada, U.S, Israel, Russia, Ukraine, India). He is an advocate of microservice based architectures, fascinated with machine learning and always eager to broaden his areas of expertise by both learning and sharing his knowledge.

Content

The prepared workshop content covers the following

Deep Learning

  • Computational Linear Algebra
  • The setup: parameters and functions, weights, biases, non-linearities, activation functions
  • Convolution and its applications
  • Matrix factorization and dimensionality reduction
  • Optimization (gradient descent, stochastic gradient descent, batch SDG, KL divergence)

MLOps

  • Environment setup (AWS, Conda, Github, Docker)
  • Overview of MLOps (Packaging, Serving, Infrastructure, Operational Considerations)
  • Cortex.dev demo
  • Model packaging - High-level overview
  • Introduction to Pickle, MLflow and ONNX
  • Docker - High-level overview
  • Creating Docker images and containers
  • Introduction to API’s (REST)
  • Introduction to Flask
  • Serve a model through Flask
  • Containerize model / API
  • Host on EC2 instance
  • Host on Fargate
  • Model training pipelines

Discount Codes

There are discount codes available

  • AISC members receive discounts according to their participation credit; log into your account (or create a new account if you haven't) to see all the available member discounts: https://members.ai.science/

Need more information?

Send us any questions you might have (events@ai.science)

Terms of Service

By signing up for this event you agree to the following terms, privacy policy, and code of conduct: https://ai.science/terms

Share with friends

Refund Policy

Contact the organizer to request a refund.

Eventbrite's fee is nonrefundable.

Save This Event

Event Saved