cli-gen >> {sessioncheck}; geo-sptemp >> vario eq; reg-other >> {ZeroOnedists}; vis-gen >> fraunces ex

Author: ercbk

qmd/cli-general.qmd

@@ -44,6 +44,29 @@
 
 -   Misc
 
+    -   Packages
+
+        -   [{]{style="color: #990000"}[fs](https://fs.r-lib.org/){style="color: #990000"}[}]{style="color: #990000"} - A cross-platform, uniform interface to file system operations
+            -   Navarro deep dive, [For fs](https://blog.djnavarro.net/posts/2024-10-06_fs/)
+        -   [{]{style="color: #990000"}[littler](https://eddelbuettel.github.io/littler/){style="color: #990000"}[}]{style="color: #990000"} - A scripting and command-line front-end for GNU R
+        -   [{]{style="color: #990000"}[ps](https://ps.r-lib.org/index.html){style="color: #990000"}[}]{style="color: #990000"} - List, Query, Manipulate System Processes
+        -   [{]{style="color: #990000"}[rapp](https://github.com/r-lib/Rapp){style="color: #990000"}[}]{style="color: #990000"} - Package for building a polished CLI application from a simple R script
+            -   An alternative front end to R, a drop-in replacement for Rscript that automatically parses command line arguments into R values
+        -   [{]{style="color: #990000"}[seekr](https://cran.r-project.org/web/packages/seekr/index.html){style="color: #990000"}[}]{style="color: #990000"} - Recursively list files from a directory, filter them using a regular expression, read their contents, and extract lines that match a user-defined pattern
+            -   Designed for quick code base exploration, log inspection, or any use case involving pattern-based file and line filtering.
+        -   [{]{style="color: #990000"}[sessioncheck](https://sessioncheck.djnavarro.net/){style="color: #990000"}[}]{style="color: #990000"} - Provides simple tools that can be called at the top of a script, and produce warnings or errors if it detects signs that the script is not being executed in a clean R session.
+            -   The intended user for sessioncheck is a beginner or intermediate level R user who needs a drop-in replacement for the simple but unsafe method of calling `rm(list = ls())` at the top of the script
+                -   Gaining the habit of restarting the R session is best when iterating or running a new script, but I guess beginning level programmers don't do this.
+            -   The only thing that `rm(list = ls())` does is remove *objects* from the global environment. This gives the user the false sense of confidence that their R session is "clean," even although in most cases it is sufficient
+                -   e.g. doesn't remove attached packages
+        -   [{]{style="color: #990000"}[sys](https://jeroen.r-universe.dev/sys){style="color: #990000"}[}]{style="color: #990000"} - Offers drop-in replacements for the `system2` with fine control and consistent behavior across platforms
+            -   Supports clean interruption, timeout, background tasks, and streaming STDIN / STDOUT / STDERR over binary or text connections.
+            -   Arguments on Windows automatically get encoded and quoted to work on different locales.
+
+    -   Resources
+
+        -   [Invoking R from the command line](https://colinfay.me/intro-to-r/appendix-b-invoking-r.html) for using `R` and `R CMD`
+
     -   The "shebang" line starting `#!` allows a script to be run directly from the command line without explicitly passing it through `Rscript` or `r`. It's not required but is a helpful convenience on Unix-like systems.
 
         ``` r
@@ -82,24 +105,6 @@
             3.  Inside folder add a batch script to launch R with --no-save arg
             4.  Add folder to path
 
--   Resources
-
-    -   [Invoking R from the command line](https://colinfay.me/intro-to-r/appendix-b-invoking-r.html) for using `R` and `R CMD`
-
--   Packages
-
-    -   [{]{style="color: #990000"}[ps](https://ps.r-lib.org/index.html){style="color: #990000"}[}]{style="color: #990000"} - List, Query, Manipulate System Processes
-    -   [{]{style="color: #990000"}[fs](https://fs.r-lib.org/){style="color: #990000"}[}]{style="color: #990000"} - A cross-platform, uniform interface to file system operations
-        -   Navarro deep dive, [For fs](https://blog.djnavarro.net/posts/2024-10-06_fs/)
-    -   [{]{style="color: #990000"}[littler](https://eddelbuettel.github.io/littler/){style="color: #990000"}[}]{style="color: #990000"} - A scripting and command-line front-end for GNU R
-    -   [{]{style="color: #990000"}[rapp](https://github.com/r-lib/Rapp){style="color: #990000"}[}]{style="color: #990000"} - Package for building a polished CLI application from a simple R script
-        -   An alternative front end to R, a drop-in replacement for Rscript that automatically parses command line arguments into R values
-    -   [{]{style="color: #990000"}[sys](https://jeroen.r-universe.dev/sys){style="color: #990000"}[}]{style="color: #990000"} - Offers drop-in replacements for the `system2` with fine control and consistent behavior across platforms
-        -   Supports clean interruption, timeout, background tasks, and streaming STDIN / STDOUT / STDERR over binary or text connections.
-        -   Arguments on Windows automatically get encoded and quoted to work on different locales.
-    -   [{]{style="color: #990000"}[seekr](https://cran.r-project.org/web/packages/seekr/index.html){style="color: #990000"}[}]{style="color: #990000"} - Recursively list files from a directory, filter them using a regular expression, read their contents, and extract lines that match a user-defined pattern
-        -   Designed for quick code base exploration, log inspection, or any use case involving pattern-based file and line filtering.
-
 -   File Paths
 
     -   Various functions to make file paths easier to work with ([source](https://bsky.app/profile/randvegan.bsky.social/post/3lllsqh33b22p))
@@ -357,7 +362,7 @@
 -   Linux utilities through Python in CLI\
 
     | Command | Purpose | More |
-    |----|----|----|
+    |------------------------|------------------------|------------------------|
     | `python3.12 -m uuid` | Like `uuidgen` CLI utility | [Docs](https://docs.python.org/3/library/uuid.html#command-line-usage) |
     | `python3.12 -m sqlite3` | Like `sqlite3` CLI utility | [Docs](https://docs.python.org/3/library/sqlite3.html#command-line-interface) |
     | `python -m zipfile` | Like `zip` & `unzip` CLI utilities | [Docs](https://docs.python.org/3/library/zipfile.html#command-line-interface) |
@@ -375,7 +380,7 @@
 -   Code Utilities\
 
     | Command | Purpose | More |
-    |----|----|----|
+    |------------------------|------------------------|------------------------|
     | `python -m pip` | Install third-party Python packages | [Docs](https://docs.python.org/3/installing/index.html) |
     | `python -m venv` | Create a virtual environment | [Docs](https://docs.python.org/3/library/venv.html) |
     | `python -m pdb` | Run the Python Debugger | [Docs](https://docs.python.org/3/library/pdb.html) |

qmd/geospatial-spat-temp.qmd

@@ -50,6 +50,8 @@
 -   Notes from
     -   [spacetime: Spatio-Temporal Data in R](https://cran.r-project.org/web/packages/spacetime/vignettes/jss816.pdf) - It's been superceded by [{stars}]{style="color: #990000"}, but there didn't seem to be much about working with vector data in [{stars}]{style="color: #990000"} vignettes. So, it seemed like a better place to start for a beginner, and I think the concepts might transfer since the packages were created by the same people. Some packages still use sp and spacetime, so it could be useful in using with those packages
     -   [Spatio-Temporal Interpolation using gstat](https://cran.r-project.org/web/packages/gstat/vignettes/spatio-temporal-kriging.pdf)
+    -   [Spatio-Temporal Kriging in R](https://www.r-bloggers.com/2015/08/spatio-temporal-kriging-in-r/)
+        -   Also shows how to create a prediction grid of roads
 -   Recommended Workflow ([source](https://www.youtube.com/watch?v=P18CZinoKos&list=WL&index=2))
     -   Go from simple analysis to complex (analysis) to get a feel for your data first. A lot of times, you probably will not need to do a spatio-temporal analysis at all.
     -   Steps
@@ -1660,6 +1662,14 @@
 
 ### Variogram {#sec-geo-sptemp-krig-var .unnumbered}
 
+#### Description
+
+-   Sample (or Empirical) Spatio-Temporal Variogram (Sherman 2011)\
+    $$
+    \hat \gamma(h,u) = \frac{1}{2N(h,u)} \sum_{N(h,u)} (Z(s_i, t_i) - Z(s_j, t_j))^2
+    $$
+    -   Calculates the semi-variance for every pair of points separated by distance $h$ and time $u$.
+
 #### `variogramST` {#sec-geo-sptemp-krig-var-vst .unnumbered}
 
 -   [Docs](https://cran.r-project.org/web/packages/gstat/refman/gstat.html#variogramST)
@@ -1670,6 +1680,7 @@
     -   Default: The length of the diagonal of the box spanning the data is divided by three.
 -   [tlags]{.arg-text} - Time lags to consider. Default is [0:15]{.arg-text}
     -   For [STIDF]{.arg-text} data, the argument [tunit]{.arg-text} is recommended (and only used in the case of STIDF) to set the temporal unit of the [tlags]{.arg-text}.
+        -   I think [tunit]{.arg-text} has the same type of input as `difftime` which are ["secs"]{.arg-text}, ["mins"]{.arg-text}, ["hours"]{.arg-text}, ["days"]{.arg-text}, or ["weeks"]{.arg-text}
     -   The number of time lags is multiplied by the number of spatial bins to get the total bins. So the fewer the better, if the situation allows it. Fewer might also reduce computation time.
     -   Seems like days are typical time unit is spatio-temporal data, so 15 lags might be a lot depending on what's being measured.
     -   Using the ACFs to get an idea
@@ -1866,8 +1877,6 @@
         -   Reducing tlags is a variance-control decision, not a bias decision
         -   heatmap also shows little correlation (high variance) approaching lag 5 while spatial correlation remains pretty strong even at 200 km (see stat ninja video)
 -   From [post](https://www.r-bloggers.com/2015/08/spatio-temporal-kriging-in-r/)
-    -   For some reason the first model, generic separable model gets automatically discarded even though it has the lowest error.
-
     -   Empircal variogram heatmap
 
         ``` r

qmd/r-snippets.qmd

@@ -128,22 +128,31 @@
 
     -   See Lewis's [Recode NA](https://cghlewis.github.io/data-wrangling-functions/recode-values/recode-na.html) tutorial for other options
 
-    -   dataframes
+    -   Remove rows with *any* NAs
 
-        ``` r
-        df %>% na.omit
-        df %>% filter(complete.cases(.))
-        df %>% tidyr::drop_na()
-        ```
+        -   Dataframes
+
+            ``` r
+            df %>% na.omit
+            df %>% filter(complete.cases(.))
+            df %>% tidyr::drop_na()
+            ```
+
+        -   Variables
+
+            ``` r
+            df %>% filter(!is.na(x1))
+            df %>% tidyr::drop_na(x1)
+            ```
 
-    -   variables
+    -   Remove rows will *all* NAs
 
         ``` r
-        df %>% filter(!is.na(x1))
-        df %>% tidyr::drop_na(x1)
+        df |> filter(!if_all(everything(), is.na))
+        df |> filter_out(if_all(everything(), is.na))
         ```
 
-    -   Replace NAs with 0 for certain columns
+    -   Replace NAs with 0 for certain columns (or the entire tibble w/`everything()`)
 
         ``` r
         d18 %>%
@@ -833,6 +842,15 @@
 
     as.POSIXct(num_datetime, origin = "1970-01-01", tz = "UTC")
     #> [1] "2026-02-19 10:35:00 UTC"
+
+    # lop off milliseconds when included (13 digits)
+    as.POSIXlt(as.numeric(
+      # paste converts to string
+      substr(paste(data$generation_time[1]), 
+             start = 1, 
+             stop = 10)), 
+      origin = "1970-01-01")
+    #> [1] "2011-10-14 11:14:46 CEST" 
     ```
 
 -   The previous month and its year ([source](https://hachyderm.io/@jimgar/113408078964321237))
@@ -851,6 +869,24 @@
     format(seq(Sys.Date(), length = 2, by = "-1 month")[2], "%B %Y")
     ```
 
+-   Subsetting
+
+    -   Base
+
+        ``` r
+        # rows
+        sub_data <- 
+          data[data$TIME >= as.POSIXct('2011-12-12 00:00 CET') & 
+                 data$TIME <= as.POSIXct('2011-12-14 23:00 CET'), ]
+
+        # columns
+        # dates is a vector of dates in same order as the columns of air
+        keep_cols <- dates >= as.Date("2005-01-01") &
+          dates <= as.Date("2010-12-31")
+
+        air_sub <- air[, keep_cols]
+        ```
+
 -   Intervals
 
     -   Difference between dates

qmd/regression-other.qmd

@@ -49,6 +49,7 @@
         -   Beta Regression
         -   [{]{style="color: #990000"}[gkwreg](https://cran.r-project.org/web/packages/gkwreg/index.html){style="color: #990000"}[}]{style="color: #990000"} - Implements regression models for bounded continuous data in the open interval (0,1) using the five-parameter Generalized Kumaraswamy distribution
         -   [{]{style="color: #990000"}[SelectBoost.beta](https://cran.r-project.org/web/packages/SelectBoost.beta/index.html){style="color: #990000"}[}]{style="color: #990000"} - Stability-Selection via Correlated Resampling for Beta-Regression Models
+        -   [{]{style="color: #990000"}[ZeroOneDists](https://cran.r-project.org/web/packages/ZeroOneDists/index.html){style="color: #990000"}[}]{style="color: #990000"} - Zero-one statistical distributions for estimating parameters and fitting regression models within the GAMLSS framework
     -   Outcome with 0s and 1s
         -   Extended support Beta Regression (Zeileis), aka XBX Regression
             -   [Thread](https://fosstodon.org/@zeileis/113146182718858220), [Paper](https://www.zeileis.org/news/xbx/), [{]{style="color: #990000"}[betareg](https://cran.r-project.org/web//packages/betareg/index.html){style="color: #990000"}[}]{style="color: #990000"}

qmd/visualization-general.qmd

@@ -314,10 +314,11 @@
         -   Interesting Gs
     -   [Fraunces](https://fonts.google.com/specimen/Fraunces)
         -   Header
+        -   The bold 400 is really thick and interesting as a title
         -   Stylistic Serif
             -   Letters: It's got some funky slants to ends of some letters. It's like it takes one curvy segment of a letter and straightens it into a slant. The f and j add an interesting sort of squiggliness or distortion to the font like looking through water.
             -   Numbers are very cool, especially 2:5. They have the flare of someone with nice penmenship.
-        -   [Example](https://kazuyanagimoto.com/blog/2025/0426_research_workflow/), [Example](https://ecogambler.netlify.app/blog/plant-community-dirichlet/) (w/Commissioner)
+        -   [Example](https://kazuyanagimoto.com/blog/2025/0426_research_workflow/), [Example](https://ecogambler.netlify.app/blog/plant-community-dirichlet/) (w/Commissioner), [Example](https://skills.amditis.tech/interview-transcription/) (w/Plus Jakarta Sans)
     -   IBM Plex Serif
         -   Body
         -   [Bruno Rodrigues' site](https://brodrigues.co/posts/2025-03-20-announcing_rixpress.html)