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/Capability/index.html

Module Modelkit.Capability

Published descriptions of optional behavior implemented by a component.

Required protocol behavior is not a capability: determinism, immutable specifications, typed failures, row alignment, and schema validation remain mandatory. These values describe only behavior that a generic consumer may need to select before invoking a component.

type support =
  1. | Supported
  2. | Unsupported
type prediction =
  1. | Direct
  2. | Labels
  3. | Positive_probabilities of int
  4. | Class_probabilities
type estimator = {
  1. estimator_sample_weight : support;
  2. estimator_fit_metadata : support;
  3. estimator_decision_function : support;
  4. estimator_predict_proba : support;
}
type transformer = {
  1. transformer_target : support;
  2. transformer_sample_weight : support;
  3. transformer_fit_metadata : support;
  4. transformer_transform_metadata : support;
}
type scorer = {
  1. scorer_sample_weight : support;
  2. scorer_prediction : prediction;
}
val estimator : ?sample_weight:support -> ?fit_metadata:support -> ?decision_function:support -> ?predict_proba:support -> unit -> estimator
val transformer : ?target:support -> ?sample_weight:support -> ?fit_metadata:support -> ?transform_metadata:support -> unit -> transformer
val scorer : ?sample_weight:support -> prediction:prediction -> unit -> scorer