scicloj.metamorph.ml.r-model-matrix — org.scicloj/metamorph.ml 1.5.1

R-style formula-based feature engineering and linear regression.

This namespace provides tools to leverage R's powerful formula syntax for feature engineering and linear modeling within Clojure. R formulas enable expressive specification of interactions, transformations, and categorical expansions without manual column manipulation.

Key Functions:

• r-model-matrix: Convert dataset + R formula to design matrix
• lm: Simplified linear regression using R formulas

Implementation Backends: The namespace supports multiple R execution backends:

• :ocpu Remote R via OpenCPU (cloud.opencpu.org) - no local R needed
• :renjine Java-based R implementation (https://renjin.org/)
• :clojisr Local R via clojisr (requires R installation)

Model Matrix Capabilities: R formulas handle:

• Basic features: y ~ x1 + x2
• Interactions: y ~ x1 * x2 (expands to x1 + x2 + x1:x2)
• Polynomial terms: y ~ x + I(x^2)
• Categorical encoding: Automatic dummy variable creation
• Intercept control: y ~ x - 1 (remove intercept)
• Exclusions: y ~ . - x3 (all columns except x3)

Linear Regression (lm): Combines formula-based feature engineering with OLS regression training. Returns a ready-to-use trained model for predictions.

Example Usage: (r-model-matrix iris-data "~ Sepal.Length + Sepal.Width" :renjine) (lm iris-data "Sepal.Width ~ Sepal.Length * Petal.Length" :Sepal.Width :ocpu)

Notes:

• OpenCPU backend is convenient but requires internet connectivity
• Renjin is standalone but may have some R incompatibilities
• clojisr requires a local R installation but offers full R compatibility
• Returned model matrices exclude row names and intercept columns by default

See also: scicloj.metamorph.ml.design-matrix for Clojure-native feature engineering

lm clj

(lm ds formula target-var formula-impl)

Train a linear model using an R-style formula.

This function combines R formula-based feature engineering with ordinary least squares (OLS) regression. It creates a design matrix from the input dataset using the specified R formula, then trains a linear model on the resulting features.

Parameters:

• ds A tech.ml.dataset dataset containing the input data with all variables referenced in the formula and target variable.

• formula A string containing the R formula (e.g., "y ~ x1 + x2 * x3"). The formula is interpreted by the R backend.

• target-var A keyword or string naming the target variable for regression. This variable must be present in the input dataset.

• formula-impl An implementation keyword for formula evaluation:

  • :ocpu Uses OpenCPU (cloud.opencpu.org), no local R needed
  • :renjine Uses Renjin, a Java implementation of R
  • :clojisr Uses clojisr with local R installation

Returns: A trained linear model (OLS from fastmath) ready for predictions. The model excludes the intercept column and row names from the design matrix by default.

Example: (lm iris-data "Sepal.Width ~ Sepal.Length + Petal.Length" :Sepal.Width :renjine)

r-model-matrix clj

(r-model-matrix ds r-formula impl)

Compute a model matrix from a dataset and an R-style formula.

Parameters:

• ds A tech.ml.dataset dataset representing the input data.
• r-formula A string containing the R formula to use for model matrix construction. The formua is interpreted by R itself, so should be full compatible
• impl An implementation keyword, either
  • :ocpu Uses an online service https://www.opencpu.org/api.html (server: cloud.opencpu.org)
  • :renjine Uses https://renjin.org/
  • :clojisr Uses https://github.com/scicloj/clojisr, which requires a local R installation

Returns a dataset containing the constructed design matrix. Dispatches to the appropriate backend implementation.

Returns a map with

• :model-matrix-dataset having the TMD containing the design matrix specified by r-formula
• :attributes the (R) attributes of the model.matrix object