home.social

Search

37 results for “tylermorganwall”

  1. Want dynamically-generated clickable links in the Positron console to your {targets} pipeline, across all your files, so you can immediately jump to an error or jump to a long-running target and inspect it? 🌳tarborist has your back!*

    *as long as you're okay building a custom version of Positron from source with a completely new ConsoleLinkProvider API😜

    github.com/tylermorganwall/pos

  2. 1/5‼️Gigantic new update + blog post! Introducing {skymodelr}: render 3D scenes in with realistically-lit skies at a given location, using just a latitude, longitude, and a time.

    Blog:
    tylermw.com/posts/rayverse/atm

    Github:
    github.com/tylermorganwall/sky

    Site:
    www.skymodelr.com

  3. ‼️ Big rayimage update (v0.23.0)! In addition to 'JPEG', 'PNG', 'TIFF', and 'EXR' files, `ray_read_image()` now supports loading 'TGA', 'BMP', 'PSD', 'GIF', 'HDR', 'PIC', and 'PNM' files (thanks to stb_image): It's a one-stop shop for bringing image data into R.

    GH:
    github.com/tylermorganwall/ray

  4. ‼️ Rayverse package update: rayvertex v0.14.0 released! This overhauls the toon shading algorithm to use a screen space jump fill algorithm (JFA), removing the brittle inverse mesh hack that existed up until now.

    github.com/tylermorganwall/ray

  5. RE: fosstodon.org/@terence/1165730

    I can’t emphasize how much I like the legend that also tells you the proportion of the whole the variable represents! It’s quite difficult to parse proportions with such a disconnected set, so this is a clever way of including it in a figure.

  6. ggplot(ritual) +
    geom_sf() +

    theme_c̵̤͛̌̎̀͘t̶̡̨̯̝̼͗̂͒͊̀̾̇̾͜h̸̡̰͉̼̖͉̉u̶̡̗̮̜͍̠̯̙̍l̸̞̲̓̒̒̓̓̃̔h̶̛̖̣̲̩̖͉̠͍̬̄̔̊̽͜u̸̱͛̈̽̀̅() +

    coord_noneuclidean()

  7. why would I get AI to write my documentation and examples, that's the best part.

  8. Get in, kids! An ADSB storm is coming!

  9. I released an update for 🌳tarborist last night! Along with bugfixes, this version brings more info on the depth of child nodes in the hover pane and DAG distance in the file picker. Most importantly, you can now hover over a target in your {targets} pipeline and get a link to jump right to it—even if it's in an entirely different file. Make navigating the pipeline much quicker.

  10. Update: It's not a true post until something is spinning in 3D 😀

  11. Just like Starbucks and Dunkin', even the rayverse is getting on the Protein-everything bandwagon :)

  12. 3D ggplot of parabolic great circle trajectories (i.e. ballistic-ish paths) on a non-linear Robinson projection: how do we ensure the 3D data is consistent with the underlying ggplot? Simple: we extract the coordinate transformation from the ggplot object itself!

  13. Well well well, what do we have here? Is that... real 3D data in my 3D ggplot? 🤫

  14. An important question from my latest blog post:

    "It would be nice to have a clean, customizable environment map that gives us that pristine golden-hour light without any real-world interlopers. Is this a realistic goal–or is it just a pie in the sky dream?"

    Post: tylermw.com/posts/rayverse/atm

  15. And here's the hex logo! And unlike Bluesky, Mastodon actually respects PNG transparency!

  16. 3/5 We can use this information along with solar positioning via {swephR} to generate a beautifully rendered realistic atmosphere in our and plots! Rayshader's `render_highquality()` now optionally takes lat/long + datetime inputs, so you don't even need to learn anything new!

  17. An "atmospheric" frame from an animation demonstrating the "Mahattanhenge" phenomenon in my upcoming blog post about rendering realistic atmospheres in R.

    What's fun about rendering a long animation overnight is it makes each morning a miniature Christmas: you never know what you're going to unwrap! Sometimes it's wonderful, and sometimes you forget to increment the counter so you end up rendering the same frame over and over again 😉

  18. Rendering the peak of Mount Everest from sunrise to sunset in R with rayshader and rayrender

  19. Big cool features coming to the : it only took a half year of CRAN ping pong for 3 low level infra packages with challenging configurations and a complete internal overhaul of rayimage to support proper colorspace management, all to just set the stage for the actual cool stuff :)

  20. 5/5 Focusing back on a single lat/long, here we simulate the analemma across the day in DC (here, in 30 minute increments) to see how the shape changes position across the sky/year. Basically, visualizing a 2D projected slice of 4D time/space: that's a fun !

  21. 1/5 Modeling the atmosphere at noon in Washington, DC for a full year in with the and my upcoming sky modeling package. If you follow the path of the sun, you might notice that it travels in a lopsided figure eight curve. Why? What's going on here?

  22. ‼️New update: v0.17.1 has been released, and with it comes some cool improvements:

    * R console ANSI full-color preview if supported (Positron does on my system!)
    * Pretty printing for RGBA arrays!
    * Proper gamma correction tracking and EXR support

    Site:
    www.rayimage.dev

  23. Testing the accuracy of simulating star trails with R and libopenexr, rendered in a pathtracer: lookin' pretty good so far! Differences seem to be primarily from real camera lens distortion from the wide angle lens versus a perspective camera.

  24. Procedurally rendered location/time-accurate star trails + simulated twilight atmosphere, now with star spectral → RGB color information included! It's subtle (even exaggerated here slightly), but there. All rendered in with

  25. 🤯 Rendering realistic sunsets in R
    🤯 🤯 Rendering realistic sunsets in R, with accurate stars for a given latitude/long
    🤯 🤯 🤯 Rendering realistic sunsets in R, with accurate stars for a given lat/long AND time/exposure, giving you beautiful star trails!

    your move, SAS

  26. Nothing is better than relaxing on a hot summer Friday night, sipping a cold drink, and enjoying a beautiful sunset accurately rendered (as if seen from a jet airliners cruising altitude while on a red eye) in your favorite statistical programming language 🥂

  27. Twilight support!

    One of my favorite things is finding a methods paper (here, atmospheric rendering), implementing it, thinking to myself "Man, it would be nice if the method supported XYZ!", and then finding that the paper implementing XYZ was released fairly recently. Truly surfing the crest of human knowledge!

  28. After setting up the required infrastructure on CRAN with several packages providing static libraries, I finally have some new features on the way! Shown here: dynamic sky generation for a given lat/long and time of day, based on the Hosek-Wilkie radiance model.