Introduction
Machine learning is the science (and art) of getting computers to act without being explicitly programmed. In recent years, machine learning has been the most prevalent tool that has allowed us to be able to design self-driving cars, practical speech recognition, effective web search, predict protein unfolding, and vastly improve understanding of the human genome. In this machine learning training, you will learn about the most effective machine learning techniques, and gain practice tinkering and implementing them yourself. More importantly, not only you’ll learn about the theoretical foundation of learning, but also gain the practical hands-on experience needed to quickly and powerfully apply these techniques to new real-world problems.
Objectives
To introduce students to the basic concepts and tools of Machine Learning
To develop skills of using recent machine learning software for solving practical problems
To use appropriate models for the given data
To understand how to evaluate models generated from data
Outcomes
By the end of the course/ training, the participants will be able to:
Understand the machine learning process; from problem definition to model creation to its deployment
Data handling with Pandas library in Python (handling missing data, handling categorical data, splitting data into train/ test set, and more)
Discover insights with Matplotlib data visualization
Implement various supervised and unsupervised machine learning models to problems
Assess performance of different models
Implement and deploy computer vision model in production
Prerequisite for joining course
Interest to learn and implement machine learning systems (essential)
Intro level Python (essential)
Good math foundation (optional)
Structure
Introduction to Machine Learning Systems
High level introduction to the machine learning world
Data
Data manipulation with Pandas
Data visualization with Matplotlib
Data evaluation and feature selection
Model (With case study: real world examples and data)
Problem definition
Model creation
Supervised
Classification
Classical machine learning algorithms
Neural networks
Loss function
Evaluation metric
Optimizer
Regression
Unsupervised
Model evaluation and tuning
K-fold cross validation
Selecting performance evaluation metrics
Hyperparameter tuning
Deployment
Model conversion
Model deployment
Logistics
6 weeks times
2 sessions per week
2 hours per session
Time & Location: 18:00 – 20:00 @ UBT
Total hours: 24
https://www.ubt-uni.net/en/study/professional-school/trainings/apply-online/