DS414
Practicing DS Skills in ML Competitions
Faculty
Alexander Guschin
Industrial Head of Machine Learning at Central University
Course length
Duration
Total hours
Credits
Language
Course type
Fee for single course
Fee for degree students
Skills you’ll learn
Overview
This course focuses on giving students the skills required to participate and excel in real-world ML tasks by practicing in ML competitions. The syllabus is designed to teach students how to solve ML problems going step-by-step. Students will create a framework for their solutions and test the problem by creating simple no-ML solutions, by preparing Python scripts, building Docker images, or deploying models. The course covers several topics, such as ML metrics understanding, optimization, EDA, feature generation, ensembling models, and debugging solutions.
Learning highlights
- Identify what methods/approaches would be feasible to solve the ML problem at hand.
- Quickly build a framework for solving the problem and provide a starter solution
- Apply ML algorithms and techniques to improve your solutions
- Turn your solution in the form that’s required: notebook, script, docker image or a service deployed
Course outline
15 classes
Session 1
Course introduction. Competition 1
Intro to ML competitions. Difference with industrial tasks. Starting your first competition. Making learning fun and building your portfolio.
Session 2
Competition 1
Early benchmark submissions without ML. ML metrics understanding and optimization. Setting up a framework for submissions.
Session 3
Competition 1
EDA. ML models. Feature generation.
Session 4
Competition 1
Ensembling models. Debugging and improving existing solutions. Competition closes.
Session 5
Competition 1
Solutions presentation, feedback and discussion.
Session 6
Competition 2. Competitions with mixed data.
Overview of competitions with texts, images, and other unstructured data. Starting the second competition.
Session 7
Competition 2
Natural Language Processing. Solving text-only and mixed data competitions.
Session 8
Competition 2
Computer Vision. Solving image-only and mixed data competitions.
Session 9
Competition 2
Different ways of combining multiple types of data in a single solution. Competition closes.
Session 10
Competition 2
Solutions presentation, feedback and discussion.
Session 11
Competition 3. Competitions that require more than a csv file.
Competitions that require submitting a Python script, Docker image, or model deployment. What is the difference? Starting the third competition.
Session 12
Competition 3
Building POC: deploying no-ML models. Setting up a framework for submissions.
Session 13
Competition 3
Building models while keeping in mind deployment restrictions. Replacing dummy models with real ones.
Session 14
Competition 3
How to showcase competition results to others? Building a portfolio. Competition closes.
Session 15
Competition 3
Solutions presentation, feedback and discussion.
Prerequisites
Participation in a few ML competitions in advance is welcomed, since it will introduce you to general concepts earlier and will allow you to learn more during the course itself.
Programming with Python.
Python for Data Analysis (Pandas, Numpy, Scipy, Sklearn).
Machine Learning (at least an introductory course).
Methodology
The course follows a hands-on approach, where students participate in various ML competitions throughout the syllabus. The instructor will introduce a new competition at the start of each week, and students will work on it throughout the week, building on the skills they have learned in the previous weeks. The students will also receive feedback on their solutions, which they will use to improve their work further.
Grading
Alexander Guschin is an Industrial Head of Machine Learning at Central University and a Fullstack ML Engineer. During his career, he worked with ML in various domains and at different scales, both as an Individual Contributor and as a DS/ML team lead. He built companion bots with LLM and Generative models, contributed to the MLOps SaaS platforms and open-source MLOps tools, including https://dvc.org and https://mlem.ai. He worked as a Machine Learning Engineering Lead at a startup centred on the application of machine learning in the industrial sector and Data Science Lead in Yandex.Go. As a teacher, he co-authored the "How to Win a Data Science Competition" curriculum at Coursera, online MLOps class at Karpov.Courses and taught classes about ML competitions and Production ML at Data Mining in Action, the largest offline open data science course in Russia, with over 500 students each year.
See full profileApply for this course
Practicing DS Skills in ML Competitions
by Alexander Guschin
Total hours
45 Hours
Dates
Jul 31 - Aug 18, 2023
Fee for single course
€1500
Fee for degree students
€750
How to secure your spot
Complete the form below to kickstart your application
Schedule your Harbour.Space interview
If successful, get ready to join us on campus
FAQ
Will I receive a certificate after completion?
Yes. Upon completion of the course, you will receive a certificate signed by the director of the program your course belonged to.
Do I need a visa?
This depends on your case. Please check with the Spanish or Thai consulate in your country of residence about visa requirements. We will do our part to provide you with the necessary documents, such as the Certificate of Enrollment.
Can I get a discount?
Yes. The easiest way to enroll in a course at a discounted price is to register for multiple courses. Registering for multiple courses will reduce the cost per individual course. Please ask the Admissions Office for more information about the other kinds of discounts we offer and what you can do to receive one.