--- title: "Projects: Software & Visualization | Matthew Oldach" url: "https://matthewoldach.com/work/" description: "Play Bloomline and explore Matthew Oldach’s complete project collection, with original publication years, software experiments, visualization and source links." --- 1. [Start](https://matthewoldach.com/) 2. Projects ![Projects chapter pop-up book](https://matthewoldach.com/models/projects/v3/poster.webp) [Start](https://matthewoldach.com/)[Services](https://matthewoldach.com/services/)[Projects](https://matthewoldach.com/work/)[Field Notes](https://matthewoldach.com/writing/)[About](https://matthewoldach.com/about/)[Contact](https://matthewoldach.com/contact/) [PuttPulse](https://matthewoldach.com/work/#puttpulse)[Shiny-filteRs](https://matthewoldach.com/work/#shiny-filters)[Hinuhinu](https://matthewoldach.com/work/#hinuhinu)Map layers: Moisture · Solar · Ocean depth [Supervised UMAP](https://matthewoldach.com/writing/supervised-umap/)[The Gamification of Fitbit](https://matthewoldach.com/writing/fitbit-gamification/)[Highlighting 3D Brain Regions](https://matthewoldach.com/writing/highlighting-3d-brain-regions/) [Work with me](https://matthewoldach.com/services/)[Experience / CV](https://matthewoldach.com/experience/) [Find the opportunity](https://matthewoldach.com/services/ai-opportunity-assessment/)[Build a focused solution](https://matthewoldach.com/services/ai-workflow-automation/)[Help the team adopt it](https://matthewoldach.com/services/ai-training-adoption/)**Good ideas deserve a way across.** Bring the ambition. Shape the plan. Build it together. Make it yours. [Write me a note](https://matthewoldach.com/contact/#enquiry)Every good collaboration starts with a note. Projects chapter preview. [Open Projects chapter page ↗](https://matthewoldach.com/work/)[Open Services page ↗](https://matthewoldach.com/services/) Choose a chapter to turn through the book in order. Viewing **Projects**. Keyboard and reduced-motion navigation switch directly. Use the chapter links to open the regular pages. The animated prototype needs JavaScript. Software · Data · Creative experiments # Projects Things I’ve built, explored and shared. Browse the complete collection, from current work to the original published experiments. [Current work](https://matthewoldach.com/work/#current-projects)[2020](https://matthewoldach.com/work/#projects-2020)[2019](https://matthewoldach.com/work/#projects-2019)[2018](https://matthewoldach.com/work/#projects-2018) Published projects are ordered newest first, using their original publication dates. 2026 / Current work ## Made to be played. [![Bloomline’s teal botanical card table, flower cards and Garden AI opponent](https://matthewoldach.com/images/bloomline/projects-preview.webp)](https://matthewoldach.com/play/bloomline/) 2026 · Playable preview ### ![Bloomline](https://matthewoldach.com/images/bloomline/brand/bloomline-wordmark.svg) A botanical card game of matching, rearranging, and outthinking Garden AI. Collect six-card Full Blooms, organize your hand, and choose your challenge. [Play Bloomline →](https://matthewoldach.com/play/bloomline/) ![Concept illustration of a disc-golf putting basket and practice area](https://matthewoldach.com/_astro/putting-area-reference-v1.iUpgWaMs_EzjrX.webp) In development ### PuttPulse A disc-golf practice product in development, shown here with its original concept artwork. [Ask about the product work →](https://matthewoldach.com/contact/) The published collection ## Ideas, in their own time. ## 2020 Jan 11, 2020 ### [UMAP Supervised Embeddings (Python->R)](https://matthewoldach.com/writing/supervised-umap/) Supervised UMAP embeddings across Python and R. [![Supervised UMAP embedding clusters with Python-to-R project artwork](https://matthewoldach.com/_astro/featured.BGJF2Knw_Z1rzKdv.webp)](https://matthewoldach.com/writing/supervised-umap/) [Read the project ↗](https://matthewoldach.com/writing/supervised-umap/) ## 2019 Nov 23, 2019 ### [Hinuhinu — Shiny development and rayshader](https://moldach.github.io/project/hinuhinu/) Hawaiian terrain, moisture, solar radiation and ocean-depth visualizations, balancing interactive ambitions with hosting constraints. Your browser does not support this video. [Open the original project clip.](https://matthewoldach.com/media/hinuhinu-walkthrough.mp4) [Open clip ↗](https://matthewoldach.com/media/hinuhinu-walkthrough.mp4) Interactive 3D maps exceeded the hosting plan’s memory. Rendered flybys retained the terrain experience while I iterated on overlays, camera movement and timing. [Read the original project ↗](https://moldach.github.io/project/hinuhinu/) Nov 23, 2019 ### [Shiny-filteRs](https://moldach.github.io/project/shiny-filters/) A mobile-oriented Shiny application combining image filters, visualization and interfaces across languages. [![Shiny-filteRs mobile screenshot showing a Voronoi-filtered portrait](https://matthewoldach.com/_astro/featured.B0pZE5pd_Z8txY8.webp)](https://moldach.github.io/project/shiny-filters/) [Read the original project ↗](https://moldach.github.io/project/shiny-filters/) Apr 21, 2019 ### [TensorFlow CNN for Fast Style Transfer](https://matthewoldach.com/writing/faststyletransfer/) Training visual styles for skiing and snowboarding footage, using Logan Engstrom’s fast-style-transfer implementation. [![Neural style-transfer rendering of a snowboarder in a brightly colored snowy landscape](https://matthewoldach.com/_astro/featured.qRi4S6-U_Z1hXEML.webp)](https://matthewoldach.com/writing/faststyletransfer/) Explore the visual experiment Content meets imagination ![Content image: riverside buildings reflected in the water](https://matthewoldach.com/_astro/example.BpkAMUM2_ZF7nt7.webp)Original ![Style-transfer output: the same buildings rendered in the colors and textures of a nebula](https://matthewoldach.com/_astro/example.BpkAMUM2_ZF7nt7.webp)Transformed Slide to explore 50% From the original 2019 experiment. Same content, a different visual language. [Read the method and credits](https://matthewoldach.com/writing/faststyletransfer/) I selected the style images, wrote a SLURM batch-training script, trained models on an HPC system, and selected outputs for the video. The original write-up describes training 50 models and choosing 20. This comparison uses the original article figure; it does not run a model in your browser. [Read the project ↗](https://matthewoldach.com/writing/faststyletransfer/) Jan 31, 2019 ### [Web Scraping Google Sheets with RSelenium](https://towardsdatascience.com/web-scraping-google-sheets-with-rselenium-9001eda399b0) Web scraping Google Sheets using browser automation with RSelenium. [![Dew-covered spiderwebs, the original featured photograph for the web-scraping project](https://matthewoldach.com/_astro/featured.CnlB3rfR_Z2723i.webp)](https://towardsdatascience.com/web-scraping-google-sheets-with-rselenium-9001eda399b0) [Read the original project ↗](https://towardsdatascience.com/web-scraping-google-sheets-with-rselenium-9001eda399b0) Jan 30, 2019 ### [10 Tips for Choosing the Optimal Number of Clusters](https://towardsdatascience.com/10-tips-for-choosing-the-optimal-number-of-clusters-277e93d72d92) Ten approaches for choosing a useful number of clusters in a dataset. [![Laptop and desktop monitor displaying code, the original clustering article photograph](https://matthewoldach.com/_astro/featured.DRMpfihu_uT50W.webp)](https://towardsdatascience.com/10-tips-for-choosing-the-optimal-number-of-clusters-277e93d72d92) [Read the original project ↗](https://towardsdatascience.com/10-tips-for-choosing-the-optimal-number-of-clusters-277e93d72d92) ## 2018 Nov 10, 2018 ### [The Gamification of Fitbit: How An API Provided The Next Level of tRaining](https://matthewoldach.com/writing/fitbit-gamification/) Accessing Fitbit API data and exploring the gamification of personal training. [![Game Boy-style Pokémon scene captioned Gaming The Fitbit API](https://matthewoldach.com/_astro/featured.qIbulrdU_Z1YICFL.webp)](https://matthewoldach.com/writing/fitbit-gamification/) [Read the project ↗](https://matthewoldach.com/writing/fitbit-gamification/) Oct 31, 2018 ### [How to Highlight 3D Brain Regions](https://matthewoldach.com/writing/highlighting-3d-brain-regions/) Three approaches to highlighting regions of interest in 3D brain images. [![Translucent 3D brain with a region highlighted in red](https://matthewoldach.com/_astro/featured.dcGZhJVU_1E4BsI.webp)](https://matthewoldach.com/writing/highlighting-3d-brain-regions/) [Read the project ↗](https://matthewoldach.com/writing/highlighting-3d-brain-regions/) Oct 31, 2018 ### [Transcriptome Assembly and Annotation](https://github.com/moldach/Transcriptome_Assembly-Annotation#transcriptome-assembly-and-annotation-pipeline) Transcriptome Assembly and Annotation. [![Sailboat jigsaw puzzle with scattered pieces, the original transcriptome assembly artwork](https://matthewoldach.com/_astro/featured.b84la_1F_ZeNXDV.webp)](https://github.com/moldach/Transcriptome_Assembly-Annotation#transcriptome-assembly-and-annotation-pipeline) [Explore the repository ↗](https://github.com/moldach/Transcriptome_Assembly-Annotation#transcriptome-assembly-and-annotation-pipeline) Oct 31, 2018 ### [vapoRwave ggplot2 themes & palettes](https://github.com/moldach/vapoRwave) This package provides a number of ggplot2 themes inspired by vaporwave, both a subgenre of electronic music and an art movement. [![Pink United States map and Andrew Yang portrait demonstrating the vapoRwave palette](https://matthewoldach.com/_astro/featured.DJzdSwYE_Z2tlTtb.webp)](https://github.com/moldach/vapoRwave) [Explore the repository ↗](https://github.com/moldach/vapoRwave) Oct 6, 2018 ### [How to Choose the Best Open Source Software](https://towardsdatascience.com/how-to-choose-the-best-open-source-software-b1cbbe4f6398) How to Choose the Best Open Source Software [![GitLab logo and development interface, the original open-source software article photograph](https://matthewoldach.com/_astro/featured.CFdDU-cb_Z2qgX9Q.webp)](https://towardsdatascience.com/how-to-choose-the-best-open-source-software-b1cbbe4f6398) [Read the original project ↗](https://towardsdatascience.com/how-to-choose-the-best-open-source-software-b1cbbe4f6398) Oct 6, 2018 ### [Reversing the order of axis in a ggplot2 scatterplot](https://medium.com/@MattOldach_65321/reversing-the-order-of-axis-in-a-ggplot2-scatterplot-b1d729b89118) Playing around with the axis display in ggplot2 [![ggplot2 bubble chart of Star Wars characters with reversed name ordering](https://matthewoldach.com/_astro/featured.YNuCzRgx_1JTzQO.webp)](https://medium.com/@MattOldach_65321/reversing-the-order-of-axis-in-a-ggplot2-scatterplot-b1d729b89118) [Read the original project ↗](https://medium.com/@MattOldach_65321/reversing-the-order-of-axis-in-a-ggplot2-scatterplot-b1d729b89118) Oct 6, 2018 ### [The Gold Standard for Data Science Project Management](https://towardsdatascience.com/the-gold-standard-of-data-science-project-management-13d68c9e85d6) Gold Standard workflow for setting up a new data science project directory and file naming [![People planning a project around a table with laptops, notebooks and printed charts](https://matthewoldach.com/_astro/featured.D9kk6pnM_Z2gAyr3.webp)](https://towardsdatascience.com/the-gold-standard-of-data-science-project-management-13d68c9e85d6) [Read the original project ↗](https://towardsdatascience.com/the-gold-standard-of-data-science-project-management-13d68c9e85d6) Jun 6, 2018 ### [Recreating data visualizations from the book Knowledge is Beautiful](https://towardsdatascience.com/recreating-data-visualizations-from-the-book-knowledge-is-beautiful-e455e7126071) In this series I will set out to recreate some of the visualization from the book “Knowledge is Beautiful” by David McCandless in R. [![Multicolored Sankey diagram recreating a Knowledge is Beautiful visualization](https://matthewoldach.com/_astro/featured.B7OjDNq0_2pARu5.webp)](https://towardsdatascience.com/recreating-data-visualizations-from-the-book-knowledge-is-beautiful-e455e7126071) [Read the original project ↗](https://towardsdatascience.com/recreating-data-visualizations-from-the-book-knowledge-is-beautiful-e455e7126071) May 7, 2018 ### [Analyzing 12 years of the Free Ride World Tour](https://towardsdatascience.com/analyzing-extreme-skiing-and-snowboarding-in-r-freeride-world-tour-1996-2018-ffde401fb3ae) Using the TwitteR package and web-scraping the FWT website for better insights [![World map showing geotagged Freeride World Tour Twitter posts](https://matthewoldach.com/_astro/featured.Dx7MAVg7_Z1ee4ny.webp)](https://towardsdatascience.com/analyzing-extreme-skiing-and-snowboarding-in-r-freeride-world-tour-1996-2018-ffde401fb3ae) [Read the original project ↗](https://towardsdatascience.com/analyzing-extreme-skiing-and-snowboarding-in-r-freeride-world-tour-1996-2018-ffde401fb3ae) Mar 10, 2018 ### [Gene set enrichment in R](https://github.com/moldach/gene-set-enrichment-R) Techniques for the analysis of gene set enrichments, pathway analysis, gene ontologies, functional analysis of metabolomic profiling and coexpression networks [![Molecular illustration of a DNA double helix](https://matthewoldach.com/_astro/featured.BkHasPaq_Z1aikzH.webp)](https://github.com/moldach/gene-set-enrichment-R) [Explore the repository ↗](https://github.com/moldach/gene-set-enrichment-R) Dates come from the original project entries, rather than the website’s later rebuild date. ## From an idea to something useful. These projects show how I approach software, data and interfaces. If you have a problem worth exploring, let’s talk. [Explore the consulting engagements ↗](https://matthewoldach.com/services/)