package modelkit

  1. Overview
  2. Docs
Portable classical machine learning workflows for OCaml

Install 

dune-project
 Dependency

github.com Readme Changelog Edit opam file Versions (4)

Authors

  1. A
    Asara

Maintainers

  1. A
    Asara developers <devs@asara.io>

Sources

modelkit-0.5.0.tbz
sha256=1fe8fa7c7f904dd098a21a2ca30fd69230750b8cf8aa2ae481b97a16531b47b4
sha512=c946cd1ac014726f4d21791e14d806a205680e6edfa2f80ed8d3680f24f48ca3c5a2f89d5afa68c11eac4053ead448b6381fb3d06dc8ff129097af6c69b262fb

doc/modelkit/Modelkit/module-type-METADATA_TRANSFORMER/index.html

Module type Modelkit.METADATA_TRANSFORMERSource

Transformer with explicit per-method metadata requests. Requests are read from the specification when packaged and remain fixed for its fitted lifetime. Both fit and transform requests are validated before training begins, since fitting also transforms the training rows. The implementation must preserve row count and order and must not retain metadata solely to substitute it for future inference inputs.

include SPECIFICATION
Sourcetype t
Sourcetype params
Sourceval clone : t -> t
Sourceval params : t -> params
Sourcetype target
Sourcetype fitted
Sourcetype rng
Sourceval fit_request : t -> Metadata.Request.t
Sourceval transform_request : t -> Metadata.Request.t
Sourceval fit : t -> metadata:Metadata.t -> rng:rng -> feature_schema:Feature_schema.t -> x:Matrix.t -> y:target option -> unit -> (fitted, Error.t) result
Sourceval transform : fitted -> metadata:Metadata.t -> feature_schema:Feature_schema.t -> x:Matrix.t -> (Matrix.t, Error.t) result
Sourceval fitted_params : fitted -> params
Sourceval input_schema : fitted -> Feature_schema.t
Sourceval output_schema : fitted -> Feature_schema.t