package caisar

You can search for identifiers within the package.

in-package search v0.2.0

A platform for characterizing the safety and robustness of artificial intelligence based software

Install

Dune Dependency

git.frama-c.com Readme Changelog Edit opam file Versions (7)

Authors

A

AISER team, Software Safety and Security Laboratory, CEA-List

Maintainers

A

AISER team, Software Safety and Security Laboratory, CEA-List

Sources

caisar-2.1.tbz

sha256=1b25c8668d428bcfc83c95147b6e45ff0a3bfa05ecd11369d12e963e29819e2e

sha512=edc7d7c0e96802811de3cb1caa3d14cc3d867ee7310748e8188eca9246a362549545c7816c8037511931dc4b7770b5ccc11b0d03abe8843b7c4db7880bf8e1fd

doc/caisar.nir/Nir/Ngraph/index.html

Module `Nir.Ngraph`Source

Neural Intermediate Representation (NIR)

NIR is a graph describing a machine learning model control flow.

A graph is described starting from its output node.

Sourcetype t

Sourceval pp : t Fmt.t

Sourceval pp_debug : t Fmt.t

Sourceval create : Node.t -> t

Create a network from its output node.t

Sourceval output : t -> Node.t

Output node.t of the network

Sourceval nodes : t -> Node.t list

Output nodes of the network

Sourceval input_shape : t -> Shape.t

Input shape of the network

Sourceval succs : t -> Node.t -> Node.t list

successors of a node.t

Sourceval iter_vertex : (Node.t -> unit) -> t -> unit

Sourceval iter_succ : (Node.t -> unit) -> t -> Node.t -> unit

Sourceval grapheasy : t -> string

Sourcemodule GFloat : sig ... end

Respect some OcamlGraph signature

Sourcemodule Dot : sig ... end

On This Page

Neural Intermediate Representation (NIR)