1.6.14.1 A table🔗ℹ

A table names the response column, and #:predictors the predictor columns. The path has R’s sequence of λ, and R’s coef(fit, s = 0.5):

> (require glmnet/datasets)

> (define predictors '("wt" "hp" "qsec"))

> (define from-table (elnet-path mtcars "mpg" #:predictors predictors))

> (vector-length (glmnet-path-lambda from-table))

58

> (coef from-table #:lambda 0.5)

'#(33.16562014425745    -3.6829279730390896    -0.02385276160227977    0.12733703169896884)

1.6.14.2 Lists and vectors🔗ℹ

The table’s columns are vectors. A vector of rows and the response column are unnamed data:

> (define mpg (dict-ref mtcars "mpg"))

> (define columns (map (λ (name) (dict-ref mtcars name)) predictors))

> (define rows (apply vector-map vector columns))

> (vector-take rows 3)

'#(#(2.62 110.0 16.46) #(2.875 110.0 17.02) #(2.32 93.0 18.61))

> (equal? (elnet-path rows mpg) from-table)

#t

1.6.14.3 A CSV file🔗ℹ

csv-file->table reads the file into a table of vector columns:

> (require glmnet/data/csv)

> (define file (collection-file-path "mtcars.csv" "glmnet" "datasets"))

> (define cars (csv-file->table file))

> (equal? (elnet-path cars "mpg" #:predictors predictors) from-table)

#t

1.6.14.4 A matrix🔗ℹ

A math/matrix matrix has the observations as its rows, and the response can be a column matrix:

> (require (only-in math/matrix ->col-matrix matrix-shape vector*->matrix))

> (define M (vector*->matrix rows))

> (matrix-shape M)

32 3

> (equal? (elnet-path M (->col-matrix mpg)) from-table)

#t

1.6.14.5 A Polars dataframe🔗ℹ

Polars reads the file itself, and the predictors leave it in one copy:

> (require (only-in polars read-csv))

> (define df (read-csv file))

> (equal? (elnet-path df "mpg" #:predictors predictors) from-table)

#t

1.6.14.6 Formulas🔗ℹ

The table, the CSV table and the dataframe name their columns, so each also fits the path as a formula model, whose coef names the coefficients:

> (define (formula-fit data)     (formula-path (mpg . ~ . wt + hp + qsec) data))

> (define formula-from-table (formula-fit mtcars))

> (define formula-from-csv (formula-fit cars))

> (define formula-from-polars (formula-fit df))

> (coef formula-from-table #:lambda 0.5)

'(("(Intercept)" . 33.16562014425745)   ("wt" . -3.6829279730390896)   ("hp" . -0.02385276160227977)   ("qsec" . 0.12733703169896884))

> (equal? (coef formula-from-csv #:lambda 0.5) (coef formula-from-table #:lambda 0.5))

#t

> (equal? (coef formula-from-polars #:lambda 0.5) (coef formula-from-table #:lambda 0.5))

#t

1.6.14.7 A design matrix, and back🔗ℹ

Converting to a design matrix once checks the data once, for many fits (see The explicit route), and fits the same path:

> (require glmnet/data/nested glmnet/data/math glmnet/data/polars)

> (define design (table->design-matrix mtcars predictors))

> (equal? (elnet-path design mpg) from-table)

#t

The design matrix converts back to each source, and table->csv writes it as CSV:

> (define first-cars (design-matrix-select-rows design '(0 1 2)))

> (design-matrix->nested first-cars)

'((2.62 110.0 16.46) (2.875 110.0 17.02) (2.32 93.0 18.61))

> (design-matrix->matrix first-cars)

(flarray #[#[2.62 110.0 16.46] #[2.875 110.0 17.02] #[2.32 93.0 18.61]])

> (design-matrix->polars first-cars)

shape: (3, 3) ┌───────┬───────┬───────┐ │ wt    ┆ hp    ┆ qsec  │ │ ---   ┆ ---   ┆ ---   │ │ f64   ┆ f64   ┆ f64   │ ╞═══════╪═══════╪═══════╡ │ 2.62  ┆ 110.0 ┆ 16.46 │ │ 2.875 ┆ 110.0 ┆ 17.02 │ │ 2.32  ┆ 93.0  ┆ 18.61 │ └───────┴───────┴───────┘

> (table->csv first-cars)

wt,hp,qsec 2.62,110,16.46 2.875,110,17.02 2.32,93,18.61