[ Web Proxy ]
URL:
Viewing: https://diffsharp.github.io/reference/diffsharp-model-parameterdict.html [Back]  [Original]

ParameterDict (DiffSharp)

ParameterDict Type

Represents a collection of named parameters.

Constructors

Constructor Description

ParameterDict()

Full Usage: ParameterDict()

Returns: ParameterDict
Returns: ParameterDict

Instance members

Instance member Description

this.[key]

Full Usage: this.[key]

Returns: string
Returns: string

this.add parameters

Full Usage: this.add parameters

Parameters:
parameters : ParameterDict

this.add parameters

Full Usage: this.add parameters

Parameters:
parameters : (string * Parameter) list

this.add (name, parameter)

Full Usage: this.add (name, parameter)

Parameters:
name : 'a
parameter : Parameter

this.backend

Full Usage: this.backend

Returns: Backend
Returns: Backend

this.clear ()

Full Usage: this.clear ()

this.copy ()

Full Usage: this.copy ()

Returns: ParameterDict

This method discards differentiability and returns a ParameterDict containing parameters that are constant tensors.

Returns: ParameterDict

this.count

Full Usage: this.count

Returns: int
Returns: int

this.device

Full Usage: this.device

Returns: Device
Returns: Device

this.dtype

Full Usage: this.dtype

Returns: Dtype
Returns: Dtype

this.flatten ?differentiable

Full Usage: this.flatten ?differentiable

Parameters:
    ?differentiable : bool

Returns: Tensor
?differentiable : bool
Returns: Tensor

this.forwardDiff (derivatives, ?nestingTag)

Full Usage: this.forwardDiff (derivatives, ?nestingTag)

Parameters:
    derivatives : ParameterDict - The derivatives of the parameters
    ?nestingTag : uint32 - The level tag for nested differentiation. Defaults to the current global nesting level

github.png [github.png] github-hover.png [github-hover.png]

Adjust the parameters to include support for forward-mode automatic differentiation.

After this call the current parameters in this dictionary will have attached derivatives for forward mode differentiation.

derivatives : ParameterDict

The derivatives of the parameters

?nestingTag : uint32

The level tag for nested differentiation. Defaults to the current global nesting level

this.iter f

Full Usage: this.iter f

Parameters:
f : string * Parameter -> unit

this.map f

Full Usage: this.map f

Parameters:
Returns: ParameterDict
f : Parameter -> Parameter
Returns: ParameterDict

this.map f

Full Usage: this.map f

Parameters:
Returns: ParameterDict
f : string * Parameter -> string * Parameter
Returns: ParameterDict

this.move (?device, ?dtype, ?backend)

Full Usage: this.move (?device, ?dtype, ?backend)

Parameters:
?device : Device
?dtype : Dtype
?backend : Backend

this.nelement

Full Usage: this.nelement

Returns: int
Returns: int

this.noDiff ()

Full Usage: this.noDiff ()

this.reverseDiff ?nestingTag

Full Usage: this.reverseDiff ?nestingTag

Parameters:
    ?nestingTag : uint32 - The level tag for nested differentiation. Defaults to the current global nesting level

github.png [github.png] github-hover.png [github-hover.png]

Adjust the parameters to include support for reverse-mode automatic differentiation.

After this call the current parameters in this dictionary will support reverse-mode differentiation. After the completion of the corresponding reverse operation, the computed derivative will be available.

?nestingTag : uint32

The level tag for nested differentiation. Defaults to the current global nesting level

this.set (other, ?differentiable, ?strict)

Full Usage: this.set (other, ?differentiable, ?strict)

Parameters:
other : ParameterDict
?differentiable : bool
?strict : bool

this.unflatten (tensors, ?differentiable)

Full Usage: this.unflatten (tensors, ?differentiable)

Parameters:
    tensors : Tensor
    ?differentiable : bool

tensors : Tensor
?differentiable : bool

this.unflattenToNew tensors

Full Usage: this.unflattenToNew tensors

Parameters:
Returns: ParameterDict
tensors : Tensor
Returns: ParameterDict

Copyright 2021, DiffSharp Contributors.



Web Proxy Viewer  |  New URL  |  Original Page