A new community of MATLAB users and fans has been set up in Brno. The first meeting of the Brno MATLAB Meetup Group was held on November 3rd. The selected topic was Data Analytics with MATLAB and it resonated with people. More than 80 attendees decided to come and hear about the new MATLAB capabilities in this area.
The meeting was led by Humusoft application engineers, Honza Studnička and Jarda Jirkovský, and was divided into two parts.
The first part was a hands-on workshop on Machine Learning. New MATLAB tools for classification and big data handling were shown using step-by- step examples. The classification example was based on prepared financial data and the goal was to find and select the best classifier using the Classification Learner app. The Classification Learner app is a built-in MATLAB graphical user interface for classification that lets you train models to classify data using supervised machine learning. Using Classification Learner, you can perform common machine learning tasks such as interactively exploring your data, selecting features, specifying validation schemes, training models and assessing results. The second example was focused on tall arrays, a new MATLAB construct for big data handling.
After a beer and sandwich break, the second part of the meeting came with a lecture. We began with a general overview of Machine Learning capabilities in MATLAB, resuming workflows shown in previous examples. Then, the topic moved to the area of Deep Learning.
Deep learning is a branch of machine learning that uses multiple nonlinear processing layers to learn useful representations of features directly from data. The topic was focused on image classification using Convolutional Neural Networks. A convolutional neural network, or CNN, is a popular deep learning architecture. Neural networks are organised into layers consisting of a set of interconnected nodes. A CNN convolves learned features with input data, and uses 2D convolutional layers that makes this architecture well suited to processing 2D data, such as images.
The lecture finished with an overview of MATLAB Big Data capabilities and tools including Memory Mapped Variables, Datastore, GPU Computing, Parallel Computing, MapReduce algorithm, Streaming Algorithms, Image Block Processing, and others.
The meeting was closed by Libor Šeda, the organiser of the Brno MATLAB Meetup Group, who shared the future plans of the group.
We would like to thank Kiwi.com for hosting our first Meetup in their cellar on Hlinky street.
We look forward to meeting you soon at the next group Meetup.
Jarda Jirkovský and Libor Šeda