mlops >> {ryl}

Author: ercbk

qmd/big-data.qmd

@@ -562,18 +562,13 @@
 - [{]{style="color: #990000"}[subsampling](https://cran.r-project.org/web/packages/subsampling/index.html){style="color: #990000"}[}]{style="color: #990000"} - Optimal Subsampling Methods for Statistical Models
   - A technique to balance the trade-off between computational efficiency and statistical efficiency when high compute resources are unavailable.
   - Provides optimal subsampling methods for various statistical models such as generalized linear models (GLM), softmax (multinomial) regression, rare event logistic regression and quantile regression model.
-- [{tidymodels}]{style="color: #990000"} database modelling packages
+- [{tidymodels}]{style="color: #990000"} database predicting packages
   - [{]{style="color: #990000"}[orbital](https://orbital.tidymodels.org/){style="color: #990000"}[}]{style="color: #990000"} - Enables running predictions of tidymodels workflows inside databases.
-
   - [{]{style="color: #990000"}[tidypredict](https://tidypredict.tidymodels.org/){style="color: #990000"}[}]{style="color: #990000"} - Used as the primary parser for models employed by the orbital
-
   - Articles
-
     - [Running tidymodel prediction workflows inside databases with orbital and Snowflake](https://posit.co/blog/running-tidymodel-prediction-workflows-inside-databases/)
     - [Deploying boosted tree models with Orbital](https://posit.co/blog/deploying-boosted-tree-models-with-orbital/)
-
   - Supported Models
-
     - Linear Regression - `lm()`
     - Generalized Linear model - `glm()`
     - Regularized Regression - `glmnet::glmnet`
@@ -590,23 +585,23 @@
       - `xgboost::xgb.Booster.complete()`
       - {catboost}
       - {lightgbm}
+  - Examples
+    - [Example]{.ribbon-highlight}: Basic [{tidypredict}]{style="color: #990000"}
 
-  - [Example]{.ribbon-highlight}: Basic [{tidypredict}]{style="color: #990000"}
-
-    ``` r
-    model <- partykit::ctree(mpg ~ am + cyl, data = mtcars)
-    tidypredict_fit(model)
-    #> case_when(cyl <= 4 ~ 26.6636363636364, cyl <= 6 & cyl > 4 ~ 19.7428571428571, 
-    #>     .default = 15.1)
-
-    model <- glmnet::glmnet(mtcars[, -1], mtcars$mpg, lambda = 0.01)
-    tidypredict_fit(model)
-    #> 13.0081464696679 + (cyl * -0.0773532164346008) + (disp * 0.00969507138358544) + 
-    #>     (hp * -0.0192462098902709) + (drat * 0.816753237688302) + 
-    #>     (wt * -3.41564341709663) + (qsec * 0.758580151032383) + (vs * 
-    #>     0.277874296242861) + (am * 2.47356523820533) + (gear * 0.645144527527598) + 
-    #>     (carb * -0.300886812079305)
-    ```
+      ``` r
+      model <- partykit::ctree(mpg ~ am + cyl, data = mtcars)
+      tidypredict_fit(model)
+      #> case_when(cyl <= 4 ~ 26.6636363636364, cyl <= 6 & cyl > 4 ~ 19.7428571428571, 
+      #>     .default = 15.1)
+
+      model <- glmnet::glmnet(mtcars[, -1], mtcars$mpg, lambda = 0.01)
+      tidypredict_fit(model)
+      #> 13.0081464696679 + (cyl * -0.0773532164346008) + (disp * 0.00969507138358544) + 
+      #>     (hp * -0.0192462098902709) + (drat * 0.816753237688302) + 
+      #>     (wt * -3.41564341709663) + (qsec * 0.758580151032383) + (vs * 
+      #>     0.277874296242861) + (am * 2.47356523820533) + (gear * 0.645144527527598) + 
+      #>     (carb * -0.300886812079305)
+      ```
 - [{]{style="color: #990000"}[VectorForgeML](https://cran.r-project.org/web/packages/VectorForgeML/index.html){style="color: #990000"}[}]{style="color: #990000"} - High-Performance Machine Learning Framework with C++ Acceleration
   - Decison Trees, K Means, KNN, Linear and Logistic Regression, PCA, Random Forest, Ridge Regression, and Softmax Regression
   - Min-max scaling, one hot encoding, model scoring, pipeline class

qmd/db-duckdb.qmd

@@ -1332,7 +1332,6 @@
   - Packages
 
     - [{]{style="color: #990000"}[duckspatial](https://cran.r-project.org/web/packages/duckspatial/index.html){style="color: #990000"}[}]{style="color: #990000"} - R Interface to 'DuckDB' Database with Spatial Extension
-    - [{]{style="color: #990000"}[ducksf](https://www.ekotov.pro/ducksf/){style="color: #990000"}[}]{style="color: #990000"} - Provides some alternatives to sf functions, which are implemented using duckdb and geoarrow. So far the only implemented function is areal interpolation working similar to the one fiund in ‘areal’ package
 
   - Spatial Predicates
 

qmd/ide-positron.qmd

@@ -2,11 +2,11 @@
 
 ## Extensions
 
--   [Reprex Generator for R](https://open-vsx.org/extension/jemus42/reprex-generator-for-r) - Brings the reprex package to your editor. Same functionality as reprex::reprex() or the RStudio addin to create reproducible examples for GitHub issues, Stack Overflow, or Slack.
+- [Reprex Generator for R](https://open-vsx.org/extension/jemus42/reprex-generator-for-r) - Brings the reprex package to your editor. Same functionality as reprex::reprex() or the RStudio addin to create reproducible examples for GitHub issues, Stack Overflow, or Slack.
 
 ## Shortcuts {#sec-ide-pos-short .unnumbered}
 
--   A lot of shortcuts don't work if you're in a quarto file. (e.g. ctrl + B makes text bold instead of unhiding primary side panel)
+- A lot of shortcuts don't work if you're in a quarto file. (e.g. ctrl + B makes text bold instead of unhiding primary side panel)
 
 | Shortcut | Description |
 |----|----|
@@ -17,14 +17,14 @@
 
 ## Settings {#sec-ide-pos-set .unnumbered}
 
--   Disable Preview Tabs: `"workbench.editor.enablePreview": false,`
+- Disable Preview Tabs: `"workbench.editor.enablePreview": false,`
 
 ## Workflows
 
--   New Project
-    -   [{uv}]{style="color: goldenrod"}
-        -   Click New Folder, Click Python
-        -   Enter name of folder, Choose the directory, Select git/no git
-        -   Select new environment or existing environment, Select uv, Select python version
-        -   Add packages: `uv add numpy scikit-learn polars statsmodels`
-        -   Do work
+- New Project
+  - [{uv}]{style="color: goldenrod"}
+    - Click New Folder, Click Python
+    - Enter name of folder, Choose the directory, Select git/no git
+    - Select new environment or existing environment, Select uv, Select python version
+    - Add packages: `uv add numpy scikit-learn polars statsmodels`
+    - Do work

qmd/mlops.qmd

@@ -11,6 +11,7 @@
 - [{]{style="color: #990000"}[onnx](https://cran.r-project.org/web/packages/onnx/index.html){style="color: #990000"}[}]{style="color: #990000"} - 'ONNX' provides an open source format for machine learning models through [{reticulate}]{style="color: #990000"}
 - [{]{style="color: #990000"}[onnxr](https://corymccartan.com/onnxr/){style="color: #990000"}[}]{style="color: #990000"} - Directly interfaces with ONNX runtime’s C++ API via cpp11. This uses far less memory and does not require a Python and TensorFlow installation.
 - [{]{style="color: goldenrod"}[PyMilo](https://github.com/openscilab/pymilo){style="color: goldenrod"}[}]{style="color: goldenrod"} ([JOSS](https://joss.theoj.org/papers/10.21105/joss.08858)) - Serializes models in a transparent human-readable format (e.g. JSON) that preserves end-to-end model fidelity and enables reliable, safe, and interpretable exchange
+- [{]{style="color: goldenrod"}[ryl](https://ryl-docs.pages.dev/){style="color: goldenrod"}[}]{style="color: goldenrod"} ([vs code ext](https://marketplace.visualstudio.com/items?itemName=owenlamont.ryl), [github](https://github.com/owenlamont/ryl)) - Rust-based YAML linter
 - [{]{style="color: #990000"}[S7schema](https://cran.r-project.org/web/packages/S7schema/index.html){style="color: #990000"}[}]{style="color: #990000"} - 'S7' Framework for Schema-Validated YAML Configuration. Validates YAML through a json schema file.
 - [{]{style="color: #990000"}[strata](https://asenetcky.github.io/strata/){style="color: #990000"}[}]{style="color: #990000"} - Simple project automation framework
   - Supports [{renv}]{style="color: #990000"}, [{cronR}]{style="color: #990000"}, and [{taskscheduleR}]{style="color: #990000"}