Photos created using a Python implementation of A Neural Algorithm of Artistic Style,Gatys et al.

Rick Torzynski

I have a Master of Arts Degree in Psychology from California State University, Fresno, with an emphasis on statistics, research methods, and computer applications. My undergraduate degree is in Psychology with two minors: 1) Peace and Conflict Studies, and 2) Russian.

I currently work as a Senior Web Developer for SpinSpire, a Jacksonville Drupal shop, contracted out to a large insurance company as a Drupal Developer. I also maintain a dedicated Linux server for small number of long time private clients. I have worked for over 25 years as a full stack web developer, with more emphasis on backend development.

After working for over 25 years as a successful web developer, I am delving deeply into Machine Learning/Deep Learning to combine my programming experience with my educational background in statistics and data analysis.

I like to consider myself a Renaissance man and my hobbies reflect that. I am a luthier, a maker of stringed instruments. I build mandolins, cigar box guitars, dulcitars, ukeleles and guitars. I also juggle, specializing in the diabolo. I used to be part of a renaissance group where I juggled, acted, participated in theatrical sword fighting demonstrations, played the recorder and sang. Currently, I enjoy playing bluegrass music on my mandolin and spending time with my family.

I am married, have one son (by marriage) and two dogs (a retired Greyhound and a pointer/labrador/bulldog mix).


A representation of my proficiency in each skill


Data Science

Bluegrass Mandolin Picker

This is a clone of Classic-Piano-Composer, https://github.com/Skuldur/Classical-Piano-Composer and instead of piano music, I used bluegrass mandolin songs from Mandolin Cafe TablEdit Library (converted to midi), https://www.mandolincafe.com/cgi-bin/tabledit/search?name=Bluegrass for training.


I converted the TablEdit files of the songs into midi files. I trained the model on a Linux computer running Ubuntu 18.04 with 32GB RAM and a GeForce GTX 1070 Ti graphics card. Took approximately 10 hours to train the model and 30secs to generated 2 minutes of random music.

This project allows you to train a neural network to generate midi music files that make use of a single instrument.

Pickin' #1:

Pickin' #2: