FREE ONLINE COURSE

From data to production ready AI model code.

Starts on 9th of January, 2023

A hands-on online course on writing AI Code

Do you want your AI team to be better and faster at developing AI projects while writing production-quality code? Are you exploring a new career path in data science, or are you looking for peers with whom to work together and refine your skills in AI and coding?
This is the course for you! Even if you are from a business background, you will learn about the end-to-end AI development process to better collaborate with your technical colleagues.
In this 5-week course, we will take you through a range of lessons spanning from data wrangling to AI code optimisation. Designed by TurinTech’s researchers, the course brings together academic rigour with industry best practices.
As an added advantage, you will have the opportunity to try out TurinTech’s AI-powered code optimisation platform, evoML.

No prior knowledge of evoML is needed to complete this course.

Enroll now for FREE!

Instructors

Dr Vitali Avagyan
Data Scientist on Machine Learning
Dr Chrystalla Pavlou
Data Scientist on Feature Engineering
Dr Paul Brookes
Data Scientist on Time Series
Dr Matthew Truscott
Data Scientist on Model Evaluation
Rafail Giavrimis
Data Scientist on Optimisation
Dr Fan Wu
CSO & Co-founder

Course details

The course is 100% online and you can learn at your own pace. It starts in December 2022. You will get notified about the exact date after signing up.

Introduction to Machine Learning

Machine learning (ML) is everywhere and difficult to elude. This course aims to explore key areas in developing a machine-learning project. The course can be useful for diverse stakeholders: those involved in academic discussions, project deployments, architecting ML software, or extracting business value out of ML models. If you are in sales, you will pick up a few concepts to better articulate your next ML sales pitch. If you are a university graduate, you can pick up the right terminology to use in your first machine-learning job interview. Whoever and wherever you are, just sign up and enjoy the course.

SECTION 1
Upstream: Data Gathering, Preparation & Feature Engineering
SECTION 2
Learning: Supervised, Unsupervised, Reinforcement Learning & Time Series
SECTION 3
Building: Model Training & Evaluation
SECTION 4
Downstream: Model Deployment, Serving, Monitoring & Maintenance
LIVE WEBINAR
Hands-on: Prepare, Train, Evaluate an ML Model
QUIZ
Machine Learning Life-Cycle
Uncover Gems in your dataset with Feature Engineering

Data and models are the two central parts of any machine learning process, determining the accuracy of the results. However, data found in practice are often in a format unsuitable for the models to consume. Therefore feature engineering plays an essential role in a machine learning pipeline: it transforms the raw data into features that better represent the underlying patterns and enhance the model performance. This lesson aims to help attendees develop the necessary skills to get the most out of their data. We will cover a range of popular feature engineering techniques, from approaches to clean and encode the data to more advanced methods of feature generation and selection, leading to improved model accuracy.

SECTION 1
Introduction to Feature Engineering and Data Cleaning
SECTION 2
Concept: Selecting column encodings and transformations
SECTION 3
Concept: Feature Generation
SECTION 4
Concept: Feature Selection
LIVE WEBINAR
An end-to-end feature engineering example
QUIZ
Basics of feature engineering
Empower that model, a time-series tale

Forecasting is the central problem whenever we are planning for the future. Will this product sell? Should I take an umbrella to work? Should I buy this house? Whether the decisions are big or small, any knowledge of future events, such as changes in the weather or interest rates, can profoundly affect the choices we make in the present. However, making forecasts is also one of the most difficult tasks to accomplish, as it continues to be challenging to produce accurate, explainable and actionable predictions of the future. For this reason, forecasting has always been one of the most engaging areas within data science. In this lesson we will introduce attendees to the basics of forecasting. We will help them understand not only the best approaches to take and the pitfalls to avoid, but also how best to interpret predictions and understand the uncertainties within them.

SECTION 1
Introduction to forecasting
SECTION 2
Concept: Data splitting and common pitfalls
SECTION 3
Concept: Dealing with uncertainty
SECTION 4
Concept: Interpreting forecasts
LIVE WEBINAR
Hands-On: training models and making predictions
QUIZ
Key concepts of forecasting
Let your model speak via Explainability

Recent advances in machine learning have been phenomenal and revolutionary. Due to these breakthroughs, models that generate fantastic results are being perceived as mystical undecipherable instruments. The goal of this short lecture is to explore how we have been able to bridge the gap between human understanding and artificial intelligence. We will learn about data analysis, or how to interpret machine learning models, and then we will dive into explainability, useful tools when interpretability gets difficult. This lesson aims to provide value to anyone interested in learning about how data analytics can be used to supercharge businesses, and for budding machine learning practitioners hoping to get their foot in the industry.

SECTION 1
Interpreting and Evaluating a Model
SECTION 2
Explainability and Model Agnostic Methods
SECTION 3
Explaining Neural Network Models
LIVE WEBINAR
Using Explainability to Make Decisions
QUIZ
Explainable AI
Automatic Code Optimisation of ML Models

The prediction latency of a machine learning model is critical, especially in sectors like trading and finance. This section of the course looks at ways of maximising code quality to achieve optimal performance. We will explore some common coding mistakes that can happen during the creation and deployment of machine learning models. The lesson will also look at optimisation techniques based on frameworks such as ONNX and TVM. This lecture is targeted towards both data scientists and engineers who want to learn how to make their entire machine learning pipeline more efficient and easily deployable to a wide range of environments.

SECTION 1
Introduction to Efficient Code Design
SECTION 2
Concept: Tracking ML Code Performance
SECTION 3
Concept: Inference using the ONNX & TVM Framework
LIVE WEBINAR
Further Optimising and Deploying an NLP model
QUIZ
Code Management and Deployment

Our technology partners

These companies are our partners whose efforts support our goals.

Fast Track
your AI projects!

Enroll in our hands-on course
View Course