home.social

#mgcvchat — Public Fediverse posts

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

  1. made some updates last week to my GAM blog: adaptive smoothing, now with plots of the smoothing parameter function

    calgary.converged.yt/articles/

    big thanks to Philip Dixon who asked an interesting question!

    #rstats #mgcvchat

  2. made some updates last week to my GAM blog: adaptive smoothing, now with plots of the smoothing parameter function

    calgary.converged.yt/articles/

    big thanks to Philip Dixon who asked an interesting question!

    #rstats #mgcvchat

  3. made some updates last week to my GAM blog: adaptive smoothing, now with plots of the smoothing parameter function

    calgary.converged.yt/articles/

    big thanks to Philip Dixon who asked an interesting question!

    #rstats #mgcvchat

  4. made some updates last week to my GAM blog: adaptive smoothing, now with plots of the smoothing parameter function

    calgary.converged.yt/articles/

    big thanks to Philip Dixon who asked an interesting question!

    #rstats #mgcvchat

  5. made some updates last week to my GAM blog: adaptive smoothing, now with plots of the smoothing parameter function

    calgary.converged.yt/articles/

    big thanks to Philip Dixon who asked an interesting question!

    #rstats #mgcvchat

  6. 📈 Yes you can do that in mgcv update

    big thanks to Zachary Susswein for spotting that my code was out of date in my neighbourhood cross-validation examples: calgary.converged.yt/articles/ calgary.converged.yt/articles/

    They are now up-to-date, as is the helper package mgcvUtils: github.com/dill/mgcvUtils

    #mgcvchat #mgcv

  7. 📈 Yes you can do that in mgcv update

    big thanks to Zachary Susswein for spotting that my code was out of date in my neighbourhood cross-validation examples: calgary.converged.yt/articles/ calgary.converged.yt/articles/

    They are now up-to-date, as is the helper package mgcvUtils: github.com/dill/mgcvUtils

    #mgcvchat #mgcv

  8. 📈 Yes you can do that in mgcv update

    big thanks to Zachary Susswein for spotting that my code was out of date in my neighbourhood cross-validation examples: calgary.converged.yt/articles/ calgary.converged.yt/articles/

    They are now up-to-date, as is the helper package mgcvUtils: github.com/dill/mgcvUtils

    #mgcvchat #mgcv

  9. 📈 Yes you can do that in mgcv update

    big thanks to Zachary Susswein for spotting that my code was out of date in my neighbourhood cross-validation examples: calgary.converged.yt/articles/ calgary.converged.yt/articles/

    They are now up-to-date, as is the helper package mgcvUtils: github.com/dill/mgcvUtils

    #mgcvchat #mgcv

  10. 📈 Yes you can do that in mgcv update

    big thanks to Zachary Susswein for spotting that my code was out of date in my neighbourhood cross-validation examples: calgary.converged.yt/articles/ calgary.converged.yt/articles/

    They are now up-to-date, as is the helper package mgcvUtils: github.com/dill/mgcvUtils

    #mgcvchat #mgcv

  11. new (out for a while but sitting in my browser from before Christmas) paper in Biometrika from Benjamin Säfken, Thomas Kneib and Simon Wood on smoothing parameter degrees of freedom

    Green OA @ Edinburgh pure.ed.ac.uk/ws/portalfiles/p

    #mgcvchat #mgcv

  12. new (out for a while but sitting in my browser from before Christmas) paper in Biometrika from Benjamin Säfken, Thomas Kneib and Simon Wood on smoothing parameter degrees of freedom

    Green OA @ Edinburgh pure.ed.ac.uk/ws/portalfiles/p

    #mgcvchat #mgcv

  13. new (out for a while but sitting in my browser from before Christmas) paper in Biometrika from Benjamin Säfken, Thomas Kneib and Simon Wood on smoothing parameter degrees of freedom

    Green OA @ Edinburgh pure.ed.ac.uk/ws/portalfiles/p

    #mgcvchat #mgcv

  14. new (out for a while but sitting in my browser from before Christmas) paper in Biometrika from Benjamin Säfken, Thomas Kneib and Simon Wood on smoothing parameter degrees of freedom

    Green OA @ Edinburgh pure.ed.ac.uk/ws/portalfiles/p

    #mgcvchat #mgcv

  15. #mgcv mini-lifehack:

    (assuming you have multithreading enabled) you can get a rough idea of what's happening when fitting a big model by looking at your CPU usage. If only 1 core is being used, the model is still "building" (assembling of design/penalty matrices), once you switch to all cores, then you're actually fitting the model. Sometimes that first model construction phase can take a long time (with a very big model), so it'll probably take a very very long time to fit. So buckle-up.

    #mgcvchat

  16. #mgcv mini-lifehack:

    (assuming you have multithreading enabled) you can get a rough idea of what's happening when fitting a big model by looking at your CPU usage. If only 1 core is being used, the model is still "building" (assembling of design/penalty matrices), once you switch to all cores, then you're actually fitting the model. Sometimes that first model construction phase can take a long time (with a very big model), so it'll probably take a very very long time to fit. So buckle-up.

    #mgcvchat

  17. #mgcv mini-lifehack:

    (assuming you have multithreading enabled) you can get a rough idea of what's happening when fitting a big model by looking at your CPU usage. If only 1 core is being used, the model is still "building" (assembling of design/penalty matrices), once you switch to all cores, then you're actually fitting the model. Sometimes that first model construction phase can take a long time (with a very big model), so it'll probably take a very very long time to fit. So buckle-up.

    #mgcvchat

  18. #mgcv mini-lifehack:

    (assuming you have multithreading enabled) you can get a rough idea of what's happening when fitting a big model by looking at your CPU usage. If only 1 core is being used, the model is still "building" (assembling of design/penalty matrices), once you switch to all cores, then you're actually fitting the model. Sometimes that first model construction phase can take a long time (with a very big model), so it'll probably take a very very long time to fit. So buckle-up.

    #mgcvchat

  19. #mgcv mini-lifehack:

    (assuming you have multithreading enabled) you can get a rough idea of what's happening when fitting a big model by looking at your CPU usage. If only 1 core is being used, the model is still "building" (assembling of design/penalty matrices), once you switch to all cores, then you're actually fitting the model. Sometimes that first model construction phase can take a long time (with a very big model), so it'll probably take a very very long time to fit. So buckle-up.

    #mgcvchat

  20. my mgcv Wrapped 2024

    top 5 basis functions:

    1. thin-plate regression splines
    2. B-splines
    3. soap film smoother
    4. cubic cyclic splines
    5. random effects (psych!)

    #mgcvchat

  21. my mgcv Wrapped 2024

    top 5 basis functions:

    1. thin-plate regression splines
    2. B-splines
    3. soap film smoother
    4. cubic cyclic splines
    5. random effects (psych!)

    #mgcvchat

  22. my mgcv Wrapped 2024

    top 5 basis functions:

    1. thin-plate regression splines
    2. B-splines
    3. soap film smoother
    4. cubic cyclic splines
    5. random effects (psych!)

    #mgcvchat

  23. my mgcv Wrapped 2024

    top 5 basis functions:

    1. thin-plate regression splines
    2. B-splines
    3. soap film smoother
    4. cubic cyclic splines
    5. random effects (psych!)

    #mgcvchat

  24. my mgcv Wrapped 2024

    top 5 basis functions:

    1. thin-plate regression splines
    2. B-splines
    3. soap film smoother
    4. cubic cyclic splines
    5. random effects (psych!)

    #mgcvchat

  25. spending some more time thinking about neighbourhood cross-validation in #mgcv (see original post here: calgary.converged.yt/articles/), but for time series.

    Pretty nice to be able to get back to a yearly trend here without needing to specify an autoregressive structure. We just need to specify a cross-validation scheme and the autocorrelation is "dealt with" during fitting.

    Full post on this soon. #mgcvchat #rstats

  26. spending some more time thinking about neighbourhood cross-validation in #mgcv (see original post here: calgary.converged.yt/articles/), but for time series.

    Pretty nice to be able to get back to a yearly trend here without needing to specify an autoregressive structure. We just need to specify a cross-validation scheme and the autocorrelation is "dealt with" during fitting.

    Full post on this soon. #mgcvchat #rstats

  27. spending some more time thinking about neighbourhood cross-validation in #mgcv (see original post here: calgary.converged.yt/articles/), but for time series.

    Pretty nice to be able to get back to a yearly trend here without needing to specify an autoregressive structure. We just need to specify a cross-validation scheme and the autocorrelation is "dealt with" during fitting.

    Full post on this soon. #mgcvchat #rstats

  28. spending some more time thinking about neighbourhood cross-validation in #mgcv (see original post here: calgary.converged.yt/articles/), but for time series.

    Pretty nice to be able to get back to a yearly trend here without needing to specify an autoregressive structure. We just need to specify a cross-validation scheme and the autocorrelation is "dealt with" during fitting.

    Full post on this soon. #mgcvchat #rstats

  29. spending some more time thinking about neighbourhood cross-validation in #mgcv (see original post here: calgary.converged.yt/articles/), but for time series.

    Pretty nice to be able to get back to a yearly trend here without needing to specify an autoregressive structure. We just need to specify a cross-validation scheme and the autocorrelation is "dealt with" during fitting.

    Full post on this soon. #mgcvchat #rstats

  30. I've been writing-up some bits on un/under-documented parts of mgcv. Here's a bit of chat about the new "neighbourhood cross-validation" method that was uploaded to arXiv a wee while ago: calgary.converged.yt/articles/

    More to come on this, including some details on how to setup neighbourhoods in practice.

    (Please @ me with errors/typos etc)

    #mgcvchat

  31. I've been writing-up some bits on un/under-documented parts of mgcv. Here's a bit of chat about the new "neighbourhood cross-validation" method that was uploaded to arXiv a wee while ago: calgary.converged.yt/articles/

    More to come on this, including some details on how to setup neighbourhoods in practice.

    (Please @ me with errors/typos etc)

    #mgcvchat

  32. I've been writing-up some bits on un/under-documented parts of mgcv. Here's a bit of chat about the new "neighbourhood cross-validation" method that was uploaded to arXiv a wee while ago: calgary.converged.yt/articles/

    More to come on this, including some details on how to setup neighbourhoods in practice.

    (Please @ me with errors/typos etc)

    #mgcvchat

  33. I've been writing-up some bits on un/under-documented parts of mgcv. Here's a bit of chat about the new "neighbourhood cross-validation" method that was uploaded to arXiv a wee while ago: calgary.converged.yt/articles/

    More to come on this, including some details on how to setup neighbourhoods in practice.

    (Please @ me with errors/typos etc)

    #mgcvchat

  34. I've been writing-up some bits on un/under-documented parts of mgcv. Here's a bit of chat about the new "neighbourhood cross-validation" method that was uploaded to arXiv a wee while ago: calgary.converged.yt/articles/

    More to come on this, including some details on how to setup neighbourhoods in practice.

    (Please @ me with errors/typos etc)

    #mgcvchat

  35. Preprint from Simon Wood on the new cross-validation smoothness estimation in #mgcv: arxiv.org/abs/2404.16490. It's a neat performant + data-efficient way to estimate GAMs based on complex CV splits (like spatial/temporal/phylo ones).

    See ?NCV in latest {mgcv} for examples (cran.r-universe.dev/mgcv/doc/m)

    I might write a helper to convert {rsample}/{spatialsample} objects into mgcv's funny CV indexing structure.

    #rstats #ml #tidymodels #mgcvchat @MikeMahoney218 @gavinsimpson @ericJpedersen @millerdl

  36. Preprint from Simon Wood on the new cross-validation smoothness estimation in #mgcv: arxiv.org/abs/2404.16490. It's a neat performant + data-efficient way to estimate GAMs based on complex CV splits (like spatial/temporal/phylo ones).

    See ?NCV in latest {mgcv} for examples (cran.r-universe.dev/mgcv/doc/m)

    I might write a helper to convert {rsample}/{spatialsample} objects into mgcv's funny CV indexing structure.

    #rstats #ml #tidymodels #mgcvchat @MikeMahoney218 @gavinsimpson @ericJpedersen @millerdl

  37. Preprint from Simon Wood on the new cross-validation smoothness estimation in #mgcv: arxiv.org/abs/2404.16490. It's a neat performant + data-efficient way to estimate GAMs based on complex CV splits (like spatial/temporal/phylo ones).

    See ?NCV in latest {mgcv} for examples (cran.r-universe.dev/mgcv/doc/m)

    I might write a helper to convert {rsample}/{spatialsample} objects into mgcv's funny CV indexing structure.

    #rstats #ml #tidymodels #mgcvchat @MikeMahoney218 @gavinsimpson @ericJpedersen @millerdl

  38. Preprint from Simon Wood on the new cross-validation smoothness estimation in #mgcv: arxiv.org/abs/2404.16490. It's a neat performant + data-efficient way to estimate GAMs based on complex CV splits (like spatial/temporal/phylo ones).

    See ?NCV in latest {mgcv} for examples (cran.r-universe.dev/mgcv/doc/m)

    I might write a helper to convert {rsample}/{spatialsample} objects into mgcv's funny CV indexing structure.

    #rstats #ml #tidymodels #mgcvchat @MikeMahoney218 @gavinsimpson @ericJpedersen @millerdl

  39. Preprint from Simon Wood on the new cross-validation smoothness estimation in #mgcv: arxiv.org/abs/2404.16490. It's a neat performant + data-efficient way to estimate GAMs based on complex CV splits (like spatial/temporal/phylo ones).

    See ?NCV in latest {mgcv} for examples (cran.r-universe.dev/mgcv/doc/m)

    I might write a helper to convert {rsample}/{spatialsample} objects into mgcv's funny CV indexing structure.

    #rstats #ml #tidymodels #mgcvchat @MikeMahoney218 @gavinsimpson @ericJpedersen @millerdl

  40. Treat today at Edinburgh Uni stats seminar: Emiko Dupont (Bath) talking about her new work on spatial confounding (arxiv.org/abs/2309.16861) #statschat #statistics #mgcvchat