The next scikit-learn release will allow inspecting the type and values of attributes of fitted estimators in Jupyter notebooks & example code rendered as HTML in sphinx-gallery powered project websites.
scikit-learn.org/dev/auto_exa...
A black and white declassified KH-7 GAMBIT spy satellite photograph of Ashgabat, Turkmenistan on July 16, 1966.
A black and white declassified KH-7 GAMBIT spy satellite photograph of Ashgabat, Turkmenistan on July 16, 1966.
A black and white declassified KH-7 GAMBIT spy satellite photograph of Ashgabat, Turkmenistan on July 16, 1966.
KH-7 images have the ability to scratch itches I didn't know existed.
This image of Ashgabat, Turkmenistan for example is incredible.
🆕 BLOG POST
There’s so much more to spatial downscaling today than classic LM or kriging. In my new blog post, I show how flexible and powerful GAMs can be for this task. #rstats #dataviz
➡️ dominicroye.github.io/blog/canary-...
1/ I'm starting to see more and more parallels between MOOCs and AI.
MOOCs worked in computer science because the end result of problems are either right or wrong. But they really ONLY worked in CS. They didn't work in any other field because there was no way to make assessment work at scale.
![2. Install the diagram extension in a Quarto project. Create a new Quarto project. From the terminal, run this to install the diagram extension:
quarto install extension pandoc-ext/diagram
3. Tell your document to use the extension. Add this to the YAML header of a Quarto document:
filters:
- diagram
4. Add a .tikz block to your document. Add a tikz code block to your document like this:
```{.tikz}
\begin{tikzpicture}[>={stealth}]
\node (x) at (0,0) {$X_{t}$};
\node (y) at (2,0) {$Y_{t + 1}$};
\node (z) at (1,1) {$Z$};
\path[->] (z) edge (x);
\path[->] (z) edge (y);
\path[->] (x) edge (y);
\end{tikzpicture}
```
5. Render! Render the document to PDF and you should see a diagram. Render it to HTML or .docx or Typst and it should use inkscape to convert it to SVG. Magic.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:2zcfjzyocp6kapg6jc4eacok/bafkreifut33lozgqgp3ualtpkyze47rtkjoysfginsohfykakfibqzetyu)
2. Install the diagram extension in a Quarto project. Create a new Quarto project. From the terminal, run this to install the diagram extension: quarto install extension pandoc-ext/diagram 3. Tell your document to use the extension. Add this to the YAML header of a Quarto document: filters: - diagram 4. Add a .tikz block to your document. Add a tikz code block to your document like this: ```{.tikz} \begin{tikzpicture}[>={stealth}] \node (x) at (0,0) {$X_{t}$}; \node (y) at (2,0) {$Y_{t + 1}$}; \node (z) at (1,1) {$Z$}; \path[->] (z) edge (x); \path[->] (z) edge (y); \path[->] (x) edge (y); \end{tikzpicture} ``` 5. Render! Render the document to PDF and you should see a diagram. Render it to HTML or .docx or Typst and it should use inkscape to convert it to SVG. Magic.
New blog post! After years of annoying shenanigans, I discovered that nowadays it's *really* straightforward to get tikz diagrams in #QuartoPub to automatically turn into SVGs in HTML, docx, and Typst! Install Inkscape, install a Quarto extension, and you're done. www.andrewheiss.com/blog/2026/02...
The next two days I will be using our "Happy Days" at @wheregroup to work on something new:
The *Project Checker*, a #QGIS plugin to check QGIS projects for potential issues
https://github.com/kannes/qgis_project_checker
It will detect e.g.
- Layers without […]
[Original post on norden.social]
I saw this when it was a preprint? They have a unique version of the estimand-estimator-estimates workflow, will be useful for may biologists, not just ecologists. We need more case studies for ecologists to emulate and for norms to shift so "regression and storytelling" becomes unpublishable.
Canopy Height Model
CHM with forest canopy gaps detected
Identifying #forest canopy gaps is important for understanding forest development & biological diversity. New R notebook outlines a workflow to detect canopy gaps from Canopy Height Models derived from ALS #lidar data. Available at:
github.com/OpenForest4D...
#RemoteSensing #Forestry #GoogleColab
GRASS packages for macOS + Linux are now on conda-forge. It allows for easier to install along other tools into the same environment for smoother data science workflows.
`conda install -c conda-forge grass`
https://grass.osgeo.org/download/conda/
Spatial Filters—In Search of the Virtual Electrode
www.jneurosci.org/content/46/7...
#GISchat: #OTD in 2000, NASA's Shuttle Radar Topography Mission (SRTM) wrapped up its mapping of much of Earth's surface. For many years (probably until Copernicus DEM was released a few yrs ago), SRTM's output was considered *the* best quality global DEM. 🧐 www.earthdata.nasa.gov/data/instrum...
I have been thinking about how LLM agents pose a threat to open source projects and what strategies can offer us at least some protection. Nevertheless, this is likely to remain a challenge: cusy.io/en/blog/how-...
#OpenSource #Ai #LLM #AISlop
What is a forest? Remarkably we don't have a consistent answer to that question,which is why we don't actually know how much forest there is on Earth or where it is. 🌏🧪🌲🌳
7 months ago ESA asked: can the open-source ecosystem handle Sentinel Zarr? 🛰️🗺️
We pushed early GeoZarr conventions with:
→ GDAL
→ OpenLayers
→ TiTiler/rasterio
→ STAC
→ rioxarray
No proprietary format. No lock-in. Just interoperability.
developmentseed.org/blog/2026-02-13-eopf-explorer-launch
📹 La presa de Montejaque (Málaga) alivia agua por primera vez en cien años de historia. La infraestructura, fallida porque fue construida en terreno poroso, se ha llenado en las últimas semanas por las intensas lluvias. Sigue la última hora del temporal social.elpais.com/aqw…
#statstab #479 Different ways of calculating OLS regression coefficients (in R)
Thoughts: The are many ways to skin a variable...
#rstats #regression #modelling #tutorial #r #glm #ols #coding
thomvolker.github.io/blog/2506_re...
As someone with an interest in medieval Slavic manuscripts only at all accessible through very cursory (important, but also culturally shaped) indexing work, I'm so excited by what we could turn up here. I gave a speculative talk about it in 2021 and now we're even closer.
New interactive global maps of 31 plant traits were prepared by Daniel Lusk, based on plant distribution data from GBIF and sPlot, plant trait data from TRY, and various satellite-based products.
Map viewer: global-traits.projects.earthengine.app/view/global-...
Article: doi.org/10.1038/s414...
"Ten simple rules for teaching data science": arxiv.org/abs/2602.02874
A new preprint by @minecr.bsky.social and myself. We'd love any feedback!
The book “Computer Age Statistical Inference” is a great read (and freely available!), and now there’s an R package that makes it easy to get the data for examples.
"publication systems [should] distinguish between dissemination of results & communication of ideas, and optimize them separately. Results should be in explicit, machine-readable form, while narrative text serves as an interpretive layer for human readers" www.biorxiv.org/content/10.6...
Metadata, data documentation, and provenance
russpoldrack.substack.com/p/metadata-d... the latest in my Better Code, Better Science series.
and lo, wend is born 👣 pypi.org/project/wend/
Introducing 𝐀𝐬𝐲𝐧𝐜-𝐆𝐞𝐨𝐓𝐈𝐅𝐅, a new library for reading GeoTIFFs and COGs. developmentseed.org/async-geotif...
- Fast, with a Rust core
- Uses a non-blocking thread pool for image decoding
- Integrates with NumPy, PyProj, Affine & Morecantile
- Remote data support with Obstore
- No GDAL dependency
Both questionable (e.g. p-hacking) and open (e.g. pre-registration) research practices are prevalent in education research. We sought to understand the explanations given by educational researchers for why either should or should not be used. Two teams of researchers independently analysed open-ended survey responses from 1488 education researchers on their feelings about questionable and open research practices. Despite using different analytic approaches, all of the major categorizations of participant responses were similar or related across teams. Our findings suggest that although respondents believe that questionable research practices should not be used, they conceded there are systemic reasons some use them. Similarly, although respondents generally support open practices, they noted situations in which they were not appropriate or necessary for education research. These findings can serve as a catalyst for training and policy initiatives. #MetaSci #Methodology #EduSky #AcademicSky #OpenSci
“Our findings suggest that although respondents believe that questionable research practices should not be used, they conceded there are systemic reasons some use them.”
Open Acc: doi.org/10.1098/rsos...
BSky authors: @sarahcaroleo.bsky.social, @jesse-fleming.bsky.social, @bryancook.bsky.social
I hate this. I hate that scholars and teachers are supposed to be digital fraud experts. I hate that this part of their job description is becoming larger and larger. I hate the widening distrust. I hate a culture that aggressively devalues the curiosity and humility required for ongoing learning.
a new #GDAL backend for xarray, coupled directly for Multidim *and* Classic raster (and actually works, pretty good improv this week)
github.com/mdsumner/gdx
Please try it out with your most|least favourite Zarr! If you have connection issues just ask I know enough to be dangerous #python
How to Download and Visualize NASA-ISRO NISAR Data in QGIS | Step-by-Step Tutorial
Learn how to download and visualize NASA-ISRO NISAR SAR satellite data in QGIS without coding! This complete tutorial shows you how to use the NASA Earthdata plugin for QGIS to access powerful SAR data.
www.egu.eu/news/472/ann...
New version of async-tiff: Fast, async reading of TIFFs, GeoTIFFs & COGs in Rust and Python: developmentseed.org/async-tiff
- New Array class for zero-copy data sharing between Rust/numpy
- Separate CPU-bound and IO-bound work, decoding image data in a thread pool
- Revamped metadata block caching
