home.social

#data-science — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #data-science, aggregated by home.social.

fetched live
  1. I turned a Raspberry Pi into a tiny weather intelligence powerhouse

    A Raspberry Pi Zero 2 W paired with a Sense HAT V2 runs a fully self-contained, edge-native machine learning weather station with no cloud, GPU, or heavy ML frameworks. The system uses pure NumPy implementations of Recursive Least Squares, Kalman filtering, conformal prediction, and drift detection, staying under 150 MB RAM. The station continuously updates its own statistical model of local atmospheric conditions, produces calibrated uncertainty intervals, and renders animated forecasts on […]

    kemal.yaylali.uk/i-turned-a-ra

  2. I turned a Raspberry Pi into a tiny weather intelligence powerhouse

    A Raspberry Pi Zero 2 W paired with a Sense HAT V2 runs a fully self-contained, edge-native machine learning weather station with no cloud, GPU, or heavy ML frameworks. The system uses pure NumPy implementations of Recursive Least Squares, Kalman filtering, conformal prediction, and drift detection, staying under 150 MB RAM. The station continuously updates its own statistical model of local atmospheric conditions, produces calibrated uncertainty intervals, and renders animated forecasts on […]

    kemal.yaylali.uk/i-turned-a-ra

  3. 🔥 The Green Illusion: Why Zoning-Level Analysis Changes the Picture

    📊 Gross community statistics can be misleading. When measuring an entire neighborhood, large municipal parks and green buffers inflate the overall 'greenness' score.

    ❗ When we filter the data down exclusively to actual Residential Land Use Districts in established communities, the picture changes significant:
    🔹 Unvegetated/Impervious surfaces (Solid) explode to ~80% (79.22%) of the total zoned area.
    🔹 Vegetative classes collapse: Functional lawns and canopy cover are compressed into narrow margins (Lawn: 2.15%, Forest: 0.59%), with community-level sparsed trees (Park) accounting for only 17.87%.

    ☝️ Moving from macro-boundaries to parcel-level zoning exposes the true physical density of Calgary’s residential footprint.

    #GIS #RemoteSensing #EarthObservation #Calgary #SpatialDataScience #LandCover #MachineLearning #Geoscience #GreennessOfCalgary #Rstats #YYC #DataScience #Urban #Sentinel

  4. 🔥 The Green Illusion: Why Zoning-Level Analysis Changes the Picture

    📊 Gross community statistics can be misleading. When measuring an entire neighborhood, large municipal parks and green buffers inflate the overall 'greenness' score.

    ❗ When we filter the data down exclusively to actual Residential Land Use Districts in established communities, the picture changes significant:
    🔹 Unvegetated/Impervious surfaces (Solid) explode to ~80% (79.22%) of the total zoned area.
    🔹 Vegetative classes collapse: Functional lawns and canopy cover are compressed into narrow margins (Lawn: 2.15%, Forest: 0.59%), with community-level sparsed trees (Park) accounting for only 17.87%.

    ☝️ Moving from macro-boundaries to parcel-level zoning exposes the true physical density of Calgary’s residential footprint.

    #GIS #RemoteSensing #EarthObservation #Calgary #SpatialDataScience #LandCover #MachineLearning #Geoscience #GreennessOfCalgary #Rstats #YYC #DataScience #Urban #Sentinel

  5. Episode 238 of the @rstats @rweekly Highlights Podcast is out! serve.podhome.fm/episodepage/r

    ✅ Reviewable R-based submissions @RConsortium
    💻 Windows ARM64 lands in R-Universe @jeroenooms
    🏫 Teaching R with {nycOpenData} (Christian Martinez) @rOpenSci

    Grab yourself a new podcast app like @podverse or @merryoscar's Fountain for an easy way to interact with the show and your hosts! newpodcastapps.com

    h/t @mike_thomas & @parmsam 🙏

    #rstats #datascience #v4v

  6. Episode 238 of the @rstats @rweekly Highlights Podcast is out! serve.podhome.fm/episodepage/r

    ✅ Reviewable R-based submissions @RConsortium
    💻 Windows ARM64 lands in R-Universe @jeroenooms
    🏫 Teaching R with {nycOpenData} (Christian Martinez) @rOpenSci

    Grab yourself a new podcast app like @podverse or @merryoscar's Fountain for an easy way to interact with the show and your hosts! newpodcastapps.com

    h/t @mike_thomas & @parmsam 🙏

    #rstats #datascience #v4v

  7. Most "big data" projects are just expensive ways to collect noise. You don't need a bigger pipe, you need a sharper question before you start filling it. A hypothesis forces you to decide what signal actually matters, then you scale that. Without it, you're just paying for storage and calling it insight. #bigdata #datascience #analytics

  8. I have created {collapse} and {fastverse} skills for claude code, as well as an experimental skill for scientific plotting in R: github.com/fastverse/skills

    If you want to write parsimonious and super-fast production-grade R code that is unlikely to break following package updates, these skills – based on package documentation and my own usage across various research codebases – now give your agents expert knowledge in operating the {fastverse}.

    Happy vibe-coding .

  9. I have created {collapse} and {fastverse} skills for claude code, as well as an experimental skill for scientific plotting in R: github.com/fastverse/skills

    If you want to write parsimonious and super-fast production-grade R code that is unlikely to break following package updates, these skills – based on package documentation and my own usage across various research codebases – now give your agents expert knowledge in operating the {fastverse}.

    Happy vibe-coding #Rstats #DataScience.