Mateusz Dorobek

Mateusz Dorobek

Data Scientist

Posts by Collection

portfolio

publications

Zastosowanie sieci neuronowych do generowania muzyki

Published in Wydział Elektroniki i Technik Infromacyjnych, Politechnika Warszawska, 2018

PL: Artykuł opisuje zastosowanie sieci neuronowych w generowaniu muzyki, zawiera także przegląd dostępnych technologii oraz opis istniejących architektur sieci neuronowych, które znalazłyby zastosowanie w tej dziedzinie. Read more

Master’s Thesis: Generating Jazz Chords Progressions Using Word Embeddings and Recurrent Neural Networks

Published in , 2020

This thesis describes the problem of generating jazz music chords using recurrent neural networks (RNN) and numerical vector representations - embeddings. To create the embedding I use techniques known from the field of natural language processing (NLP). It is possible because of certain structural similarities between spoken language and music. I study the relationship between the chords in the hidden space generated by the algorithms: Word2Vec, FastText and Multi-hot. Visualization of chords in reduced vector representation space using t-SNE and PCA algorithms illustrates a lot of dependencies between the chords derived from the principles of jazz harmony. The test results confirmed the highest performance of chord generation when using the Word2Vec model in the Skip-Gram variant. To generate chords I use their vector representation. I analyze the performance of sixteen models based on the recursive neural networks in terms of their hyper parameters and architecture. I also study the impact of the algorithm used to create the numerical representation on the results of recurrent models. To train neural networks I use a set of jazz standards obtained from publicly available Internet sources in a format allowing for the extraction of chord sequences. The experiments shows the highest performance of deep models with recurrent layers - GRU and LSTM, but the one with the shortest runtime is the model with two GRU layers, Dropout regularization layer and Dense layer. I use trained models to build a simple program capable of generating continuous chord sequences in the unsupervised variant and with partially controlling this generation by the user. For implementation of all algorithms, visualization and data processing I use Python language and its libraries: tensorflow, keras, gensim, scikit-learn, numpy, pychord, music21 etc. Read more

talks

PL Warsztaty ML MINI PW

Published:

PL: Warsztaty z Uczenia Maszynowego na wydziale MINI Politechniki Warszawskiej - 28 maja 2019 EN: Workshop on Machine Learning on faculty of Mathematics Warsaw University of Technology - 28 may 2019 Read more

EN Jazz chord generation using Deep Learning

Published:

A presentation of my master thesis project using language models to represent Jazz Chords and Recurrent Neural Networks to generate jazz chords progressions. Event was held by Python Barcelona. Read more

Data Science Summit 2020

Published:

Wystąpiłem jako prelegent na konferencji Data Science Summit 2020. Jest to jedno z największych i najważniejszych polskich wydarzeń data science w Polsce. Agenda DSS20 jest bardzo bogata (120+ wystąpień, 15 ścieżek) - każdy znajdzie coś dla siebie (od przetwarzania, poprzez analizę i uczenie maszynowe, aż po wizualizację danych). Read more

MLinPL 2020

Published:

ML in PL 2020 odbyło się w formie online. Moim zdaniem najlepiej zorganizowane wydarzenie AI 2020 roku. Kudos dla organizatorów. Read more

teaching

Machine Learning Workshop

Machine Learning Workshops Materials.

Workshop on Machine Learning at the Faculty of Mathematics, Warsaw University of Technology - 28 May 2019. Run the notebook from workshops here: Binder Read more

Python Course

Python Beginner to Intermediate Course materials and tasks for my students.

I use Python in my everyday work as a tool for building applications, web scraping, artificial intelligence, music, and many other interesting things. My Python-Course repository is a handy database with learning materials and tasks for my students. Read more