Package 'umbridge'

Title: Integration for the UM-Bridge Protocol
Description: A convenient wrapper for the UM-Bridge protocol. UM-Bridge is a protocol designed for coupling uncertainty quantification (or statistical / optimization) software to numerical models. A model is represented as a mathematical function with optional support for derivatives via Jacobian actions etc.
Authors: Linus Seelinger [aut, cre]
Maintainer: Linus Seelinger <[email protected]>
License: MIT + file LICENSE
Version: 1.0
Built: 2025-03-08 03:30:54 UTC
Source: https://github.com/cran/umbridge

Help Index

Evaluate Hessian of model.

Description

Evaluate Hessian of model.

Usage

apply_hessian(
  url,
  name,
  out_wrt,
  in_wrt1,
  in_wrt2,
  parameters,
  sens,
  vec,
  config = jsonlite::fromJSON("{}")
)

Arguments

url

URL the model is running at.
name

Name of the desired model.
out_wrt

Output variable to take Hessian with respect to.
in_wrt1

First input variable to take Hessian with respect to.
in_wrt2

Second input variable to take Hessian with respect to.
parameters

Model input parameter (a list of vectors).
sens

Sensitivity with respect to output.
vec

Vector to multiply Hessian by.
config

Model-specific configuration options.

Value

Hessian with respect to given inputs and outputs, applied to given sensitivity and vector.

Evaluate Jacobian of model.

Description

Evaluate Jacobian of model.

Usage

apply_jacobian(
  url,
  name,
  out_wrt,
  in_wrt,
  parameters,
  vec,
  config = jsonlite::fromJSON("{}")
)

Arguments

url

URL the model is running at.
name

Name of the desired model.
out_wrt

Output variable to take Jacobian with respect to.
in_wrt

Input variable to take Jacobian with respect to.
parameters

Model input parameter (a list of vectors).
vec

Vector to multiply Jacobian by.
config

Model-specific configuration options.

Value

Jacobian with respect to given input and output variables, applied to given vector.

Evaluate model.

Description

Evaluate model.

Usage

evaluate(url, name, parameters, config = jsonlite::fromJSON("{}"))

Arguments

url

URL the model is running at.
name

Name of the desired model.
parameters

Model input parameter (a list of vectors).
config

Model-specific configuration options.

Value

The model output (a list of vectors).

Get models supported by server.

Description

Get models supported by server.

Usage

get_models(url)

Arguments

url

URL the model is running at.

Value

List of models supported by server.

Evaluate gradient of target functional depending on model.

Description

Evaluate gradient of target functional depending on model.

Usage

gradient(
  url,
  name,
  out_wrt,
  in_wrt,
  parameters,
  sens,
  config = jsonlite::fromJSON("{}")
)

Arguments

url

URL the model is running at.
name

Name of the desired model.
out_wrt

Output variable to take gradient with respect to.
in_wrt

Input variable to take gradient with respect to.
parameters

Model input parameter (a list of vectors).
sens

Sensitivity of target functional with respect to model output.
config

Model-specific configuration options.

Value

Gradient of target functional.

Retrieve model's input dimensions.

Description

Retrieve model's input dimensions.

Usage

model_input_sizes(url, name, config = jsonlite::fromJSON("{}"))

Arguments

url

URL the model is running at.
name

Name of the desired model.
config

Model-specific configuration options.

Value

List of input dimensions.

Retrieve model's output dimensions.

Description

Retrieve model's output dimensions.

Usage

model_output_sizes(url, name, config = jsonlite::fromJSON("{}"))

Arguments

url

URL the model is running at.
name

Name of the desired model
config

Model-specific configuration options.

Value

List of output dimensions.

Check if model's protocol version is supported by this client.

Description

Check if model's protocol version is supported by this client.

Usage

protocol_version_supported(url)

Arguments

url

URL the model is running at.

Value

TRUE if model's protocol version is supported by this client, FALSE otherwise.

Check if model supports Hessian action.

Description

Check if model supports Hessian action.

Usage

supports_apply_hessian(url, name)

Arguments

url

URL the model is running at.
name

Name of the desired model.

Value

TRUE if model supports Hessian action, FALSE otherwise.

Check if model supports Jacobian action.

Description

Check if model supports Jacobian action.

Usage

supports_apply_jacobian(url, name)

Arguments

url

URL the model is running at.
name

Name of the desired model.

Value

TRUE if model supports Jacobian action, FALSE otherwise.

Check if model supports evaluation.

Description

Check if model supports evaluation.

Usage

supports_evaluate(url, name)

Arguments

url

URL the model is running at.
name

Name of the desired model.

Value

TRUE if model supports evaluation, FALSE otherwise.

Check if model supports gradient evaluation.

Description

Check if model supports gradient evaluation.

Usage

supports_gradient(url, name)

Arguments

url

URL the model is running at.
name

Name of the desired model.

Value

TRUE if model supports gradient evaluation, FALSE otherwise.