https://www.rostrum.blog/tags/magick/ Recent content in Magick on rostrum.blog Hugo -- gohugo.io en-gb Sun, 07 Nov 2021 00:00:00 +0000 https://www.rostrum.blog/2021/11/07/deepfry/ Sun, 07 Nov 2021 00:00:00 +0000 https://www.rostrum.blog/2021/11/07/deepfry/ tl;dr Now you can use a function to deep fry memes in R. Extra crispy You can make memes in R with packages like Guangchang Yu’s {meme}. You could even post them to Twitter with #RStatsMemes for @rstatsmemes to find. However, it’s no longer enough to present memes as-is. They must be deep-fried to become modern and ironic. It will help people think that your meme is so edgy that it’s been re-saved thousands of times. https://www.rostrum.blog/2021/10/05/gorilla/ Tue, 05 Oct 2021 00:00:00 +0000 https://www.rostrum.blog/2021/10/05/gorilla/ tl;dr You can convert a line drawing to datapoints with a sprinkle of {magick}. Ape escape Have you seen that video where you’re so focused on counting basketball passes that you fail to see the gorilla moving across the screen? This kind of selective attention was studied by two researchers, Yanai and Lercher, who provided subjects with a dataset that looked like a gorilla when plotted. The gorilla was found less often if the subjects were also given a hypothesis to investigate. https://www.rostrum.blog/2021/07/31/og-emoji-svg/ Sat, 31 Jul 2021 00:00:00 +0000 https://www.rostrum.blog/2021/07/31/og-emoji-svg/ tl;dr I wrote code to produce SVG versions of the ‘first-ever’ emoji set. Using R, I scraped Emojipedia with the {polite} package and then handled images with {png}, {magick} and {svglite}. Important archival work I posted recently on creating ‘pixel art’ in R and have since stumbled upon an old post by mikefc on the coolbutuseless blog with a method that makes it easier to convert from an image to its ‘pixels’. https://www.rostrum.blog/2021/07/23/london-colour/ Fri, 23 Jul 2021 00:00:00 +0000 https://www.rostrum.blog/2021/07/23/london-colour/ tl;dr I used the {rtweet} and {magick} R packages to fetch tweets of random satellite images of London and then reduced each one to a single representative colour. Green/grey belt I created the @londonmapbot Twitter bot to tweet out satellite images of random points in Greater London. You can read earlier posts about how it was made and how I mapped the points interactively. I figured we could sample these to get to ‘the colours of London’, which can be mapped or tiled. https://www.rostrum.blog/2021/06/28/pixel-art/ Mon, 28 Jun 2021 00:00:00 +0000 https://www.rostrum.blog/2021/06/28/pixel-art/ It’s dangerous to code alone… tl;dr You can use R’s image() function to convert a matrix to a pixelly graphic. Pixel fixation My last post was about the {emojiscape} package, which makes a little scene out of sampled emojis. Following a similar approach, you could write a matrix by hand and plot it via the base function image(). Here’s a very basic example with a ‘glider’ from Conway’s Game of Life. https://www.rostrum.blog/2018/11/25/art-of-the-possible/ Sun, 25 Nov 2018 00:00:00 +0000 https://www.rostrum.blog/2018/11/25/art-of-the-possible/ ‘Walrus rainbow vomit’ is a sentence I’d never thought I’d type (via Giphy) Note I later learnt about {colorfindr} by David Zumbach, which can extract colours from images, provide composition details and generate palettes. Check it out. tl;dr I used the {magick} package in R to map an image’s colours to their nearest match from a simplified palette, then quantified how much of the image was covered by each colour in that palette. https://www.rostrum.blog/2018/11/21/waggle-dance/ Wed, 21 Nov 2018 00:00:00 +0000 https://www.rostrum.blog/2018/11/21/waggle-dance/ A bee scene from irreverent 90s Nicktoon ‘Hey Arnold!’ (via Giphy) How to plot grouped continuous data? A boxplot lets you show continuous data split by categories, but it hides the data points and doesn’t tell you much about distribution. A violin chart will show the distribution but you still don’t know about the density of data. Stripcharts show the data for each category as individual points. The points can be layered on top of each other where they take the same Y value and can be stretched arbitrarily along the X axis.