Train a CoDaCoRe Classifier

Fits an official CoDaCoRe model when the `codacore` package is available. If the package is unavailable or the TensorFlow backend errors and `backend = "auto"`, the function falls back to a deterministic sparse-balance logistic model built from ranked CLR effects.

Usage

train_codacore(
  X_train,
  y_train,
  epochs = 30L,
  lr = 0.001,
  batch_size = 256L,
  class_weight = TRUE,
  device = NULL,
  verbose = TRUE,
  seed = 42L,
  positive = NULL,
  input_scale = c("log", "raw"),
  backend = c("auto", "codacore", "fallback"),
  max_balances = 5L,
  top_k = 3L,
  cv_folds = 5L,
  fast = TRUE
)

Arguments

X_train

Numeric training matrix with samples in rows and features in columns.

y_train

Binary outcome vector.

epochs

Number of optimization epochs. Defaults to `30`.

lr

Learning rate placeholder retained for API compatibility with the other Paper 1 learners. The fallback backend does not use it directly.

batch_size

Mini-batch size passed to the official `codacore` optimizer. Defaults to `256`.

class_weight

Logical; if `TRUE`, use inverse-frequency weighting in the fallback logistic model.

device

Included for API consistency; currently ignored.

verbose

Logical; print backend messages. Defaults to `TRUE`.

seed

Integer random seed. Defaults to `42`.

positive

Optional positive-class label for non-numeric outcomes.

input_scale

Either `"log"` (default) or `"raw"`.

backend

One of `"auto"`, `"codacore"`, or `"fallback"`.

max_balances

Maximum number of sparse balances for the fallback model and `maxBaseLearners` for the official backend.

top_k

Initial sparse balance size used by the fallback model.

cv_folds

Number of internal cross-validation folds passed to the official backend.

fast

Logical; passed through to the official `codacore` call.

Value

An object of class `"omicselector_codacore"` suitable for [predict_codacore()].

References

Gordon-Rodriguez E, Susin A, McEwen JD, et al. (2022). CoDaCoRe: learning sparse log-ratios for high-throughput sequencing data. Bioinformatics, 38(12), 3179-3186.

Aitchison J. (1986). The Statistical Analysis of Compositional Data. Chapman and Hall.

Gloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ. (2017). Microbiome datasets are compositional: and this is not optional. Frontiers in Microbiology, 8, 2224.

Examples

X_train <- matrix(rnorm(14 * 20), nrow = 20)
y_train <- factor(rep(c("control", "case"), each = 10))
fit <- train_codacore(X_train, y_train, backend = "fallback", verbose = FALSE)
fit$backend
#> [1] "fallback"