From 304a97cedbadd383a4c774b649d1a3f371c08d05 Mon Sep 17 00:00:00 2001 From: "Documenter.jl" Date: Tue, 6 Feb 2024 21:23:41 +0000 Subject: [PATCH] build based on b114ca9 --- dev/.documenter-siteinfo.json | 2 +- dev/adaptivemap/index.html | 4 ++-- dev/extras/index.html | 2 +- dev/index.html | 2 +- dev/map/index.html | 6 +++--- dev/multiindex/index.html | 4 ++-- dev/trainmap/index.html | 4 ++-- 7 files changed, 12 insertions(+), 12 deletions(-) diff --git a/dev/.documenter-siteinfo.json b/dev/.documenter-siteinfo.json index 5a8e39a..d64c461 100644 --- a/dev/.documenter-siteinfo.json +++ b/dev/.documenter-siteinfo.json @@ -1 +1 @@ -{"documenter":{"julia_version":"1.9.4","generation_timestamp":"2024-01-30T14:05:49","documenter_version":"1.2.1"}} \ No newline at end of file +{"documenter":{"julia_version":"1.9.4","generation_timestamp":"2024-02-06T21:23:38","documenter_version":"1.2.1"}} \ No newline at end of file diff --git a/dev/adaptivemap/index.html b/dev/adaptivemap/index.html index 2667139..7be904b 100644 --- a/dev/adaptivemap/index.html +++ b/dev/adaptivemap/index.html @@ -1,4 +1,4 @@ -Adaptive training · MParT.jl

Adaptive Map Training Routines

MParT.TrainMapAdaptive — Function
TrainMapAdaptive(msets, objective, options)

Implements the ATM algorithm Baptista, et al.

Takes in initial guess of multiindex sets for each output dimension and adapts those sets to better approximate the probability distribution of interest using monotone transport maps.

Examples

source
MParT.ATMOptions — Function
`ATMOptions(;kwargs...)`

Options for using the Adaptive Transport Map algorithm from Baptista, et al.

Inherits all keywords from MapOptions and TrainOptions, plus the arguments below.

Arguments

  • maxPatience::Int: Number of "stationary" algorithm iterations tolerated
  • maxSize::Int: the total number of multiindices in the entire map that the algorithm is allowed to add. Should be larger than the sum of the sizes of all multiindex sets across all dimensions for the map
  • maxDegrees::MultiIndex: The maximum degree of any expansion term for each dimension (should be length of dimensions of the map)

Examples

julia> maxDegrees = MultiIndex(2,3); # limit both dimensions by order 3
+Adaptive training · MParT.jl
Adaptive Map Training Routines
MParT.TrainMapAdaptive — Function
TrainMapAdaptive(msets, objective, options)
Implements the ATM algorithm Baptista, et al.
Takes in initial guess of multiindex sets for each output dimension and adapts those sets to better approximate the probability distribution of interest using monotone transport maps.
Examples
source
MParT.ATMOptions — Function
`ATMOptions(;kwargs...)`
Options for using the Adaptive Transport Map algorithm from Baptista, et al.
Inherits all keywords from MapOptions and TrainOptions, plus the arguments below.
Arguments
  • maxPatience::Int: Number of "stationary" algorithm iterations tolerated
  • maxSize::Int: the total number of multiindices in the entire map that the algorithm is allowed to add. Should be larger than the sum of the sizes of all multiindex sets across all dimensions for the map
  • maxDegrees::MultiIndex: The maximum degree of any expansion term for each dimension (should be length of dimensions of the map)
Examples
julia> maxDegrees = MultiIndex(2,3); # limit both dimensions by order 3
 
-julia> ATMOptions(opt_alg="LD_SLSQP", maxDegrees=maxDegrees);
See also TrainMapAdaptive, TrainOptions, MapOptions
source

+julia> ATMOptions(opt_alg="LD_SLSQP", maxDegrees=maxDegrees);

See also TrainMapAdaptive, TrainOptions, MapOptions

source
diff --git a/dev/extras/index.html b/dev/extras/index.html index bafe9fb..7a21d11 100644 --- a/dev/extras/index.html +++ b/dev/extras/index.html @@ -1,2 +1,2 @@ -Extras · MParT.jl

Routines not available via Pkg

MParT.Serialize — Function
Serialize(obj, filename)

Serializes obj into file filename. REQUIRES CEREAL INSTALLATION.

source
MParT.Deserialize — Function
Deserialize(obj, filename)

Deserializes filename and puts the contents in obj. REQUIRES CEREAL INSTALLATION.

The object obj can be of type MapOptions or FixedMultiIndexSet. This will create a new pointer– other objects with the same pointer will not be modified, but the contents of obj will now point to the deserialized object.

source
MParT.DeserializeMap — Function
DeserializeMap(filename::String)

REQUIRES CEREAL INSTALLATION. Deserializes a map and returns its input dimension, output dimension, and coefficient.

source
+Extras · MParT.jl

Routines not available via Pkg

MParT.Serialize — Function
Serialize(obj, filename)

Serializes obj into file filename. REQUIRES CEREAL INSTALLATION.

source
MParT.Deserialize — Function
Deserialize(obj, filename)

Deserializes filename and puts the contents in obj. REQUIRES CEREAL INSTALLATION.

The object obj can be of type MapOptions or FixedMultiIndexSet. This will create a new pointer– other objects with the same pointer will not be modified, but the contents of obj will now point to the deserialized object.

source
MParT.DeserializeMap — Function
DeserializeMap(filename::String)

REQUIRES CEREAL INSTALLATION. Deserializes a map and returns its input dimension, output dimension, and coefficient.

source
diff --git a/dev/index.html b/dev/index.html index fc6b7ff..bfb34e1 100644 --- a/dev/index.html +++ b/dev/index.html @@ -1,2 +1,2 @@ -Home · MParT.jl

MParT

Documentation for MParT.jl. This documentation is incomplete and only features some basic functionality. For documentation of all functions as well as source installation instructions, other binding information, and many examples, please look at the C++ documentation. If there's a function there unavailable in Julia, please let the developers know via a GitHub issue.

General Utilities

+Home · MParT.jl

MParT

Documentation for MParT.jl. This documentation is incomplete and only features some basic functionality. For documentation of all functions as well as source installation instructions, other binding information, and many examples, please look at the C++ documentation. If there's a function there unavailable in Julia, please let the developers know via a GitHub issue.

General Utilities

diff --git a/dev/map/index.html b/dev/map/index.html index 2b5a059..fdf2dce 100644 --- a/dev/map/index.html +++ b/dev/map/index.html @@ -1,5 +1,5 @@ -Map Functionality · MParT.jl

Basic map Routines

MParT.Evaluate — Function
Evaluate(map, points)

Evaluates the function map at points, where each column of points is a different sample.

If map $:\mathbb{R}^m\to\mathbb{R}^n$, then points $\in\mathbb{R}^m\times\mathbb{R}^k$, where $k$ is the number of points.

source
MParT.Inverse — Function
Inverse(map, y, x)

If map represents function $T(y,x)$, then this function calculates $T(y,\cdot)^{-1}(x)$.

If map is square, you still require y, but it can be a 0 x k matrix.

source
MParT.ComposedMap — Type
ComposedMap(maps::Vector)

Creates a ComposedMap from a vector of ConditionalMapBase objects.

source
MParT.MapOptions — Function
MapOptions(;kwargs...)

Creates options for parameterized map.

All possible keyword arguments are in example, with some important arguments described below. See C++ documentation for an exhaustive description.

Arguments

  • basisType::String: Includes "ProbabilistHermite", "PhysicistHermite", "HermiteFunctions"
  • basisLB::Float64,basisUB::Float64: The bounds for where we start linearizing the map. These default to infinities, but often making the data near the origin and setting them to a small finite number (e.g. +-3) works well.

Example

julia> MapOptions(basisType="HermiteFunctions", basisLB=-3., basisUB=3.)
+Map Functionality · MParT.jl
Basic map Routines
MParT.Evaluate — Function
Evaluate(map, points)
Evaluates the function map at points, where each column of points is a different sample.
If map $:\mathbb{R}^m\to\mathbb{R}^n$, then points $\in\mathbb{R}^m\times\mathbb{R}^k$, where $k$ is the number of points.
source
MParT.Inverse — Function
Inverse(map, y, x)
If map represents function $T(y,x)$, then this function calculates $T(y,\cdot)^{-1}(x)$.
If map is square, you still require y, but it can be a 0 x k matrix.
source
MParT.ComposedMap — Type
ComposedMap(maps::Vector)
Creates a ComposedMap from a vector of ConditionalMapBase objects.
source
MParT.MapOptions — Function
MapOptions(;kwargs...)
Creates options for parameterized map.
All possible keyword arguments are in example, with some important arguments described below. See C++ documentation for an exhaustive description.
Arguments
  • basisType::String: Includes "ProbabilistHermite", "PhysicistHermite", "HermiteFunctions"
  • basisLB::Float64,basisUB::Float64: The bounds for where we start linearizing the map. These default to infinities, but often making the data near the origin and setting them to a small finite number (e.g. +-3) works well.
Example
julia> MapOptions(basisType="HermiteFunctions", basisLB=-3., basisUB=3.)
 basisType = HermiteFunctions
 basisLB = -3
 basisUB = 3
@@ -13,7 +13,7 @@
 quadPts = 5
 contDeriv = true
 nugget = 0
-
See also CreateComponent, TriangularMap, CreateTriangular
source
MParT.TriangularMap — Type
TriangularMap(maps::Vector, move_coeffs::Bool = true)
Creates a TriangularMap from a vector of ConditionalMapBase objects.
TODO: The new object takes ownership of the coeffs of the maps in maps if move_coeffs is true.
Examples
julia> dim, order = 5, 3;
 
 julia> msets = [FixedMultiIndexSet(d, order) for d in 1:dim];
 
@@ -21,4 +21,4 @@
 
 julia> components = [CreateComponent(mset, opts) for mset in msets];
 
-julia> trimap = TriangularMap(components);
source
MParT.CreateTriangular — Function
CreateTriangular(inDim::Int, outDim::Int, p::Int, opts::MapOptions)
Creates a total order p map with dimensions inDim and outDim with specifications opts.
source
MParT.CreateComponent — Function
CreateComponent(mset::FixedMultiIndexSet, opts::MapOptions)
Create a single-output component with approximation order given by mset and specified by opts
source

+julia> trimap = TriangularMap(components);
source
MParT.CreateTriangular — Function
CreateTriangular(inDim::Int, outDim::Int, p::Int, opts::MapOptions)

Creates a total order p map with dimensions inDim and outDim with specifications opts.

source
MParT.CreateComponent — Function
CreateComponent(mset::FixedMultiIndexSet, opts::MapOptions)

Create a single-output component with approximation order given by mset and specified by opts

source
diff --git a/dev/multiindex/index.html b/dev/multiindex/index.html index 7be0c19..84e496a 100644 --- a/dev/multiindex/index.html +++ b/dev/multiindex/index.html @@ -4,8 +4,8 @@ julia> midx = MultiIndex(degrees); julia> midx[3] -0x00000001

See also MultiIndexSet, FixedMultiIndexSet, Fix

source
MParT.MultiIndexSet — Type
MultiIndexSet(A::AbstractVecOrMat{<:Integer})

Create a set of MultiIndices from the rows of A.

These indices represent a basis for a multivariate expansion or, further, monotone expansion. Each element of the set is a MultiIndex representing one basis function via the degrees in each dimension.

Example

julia> # Functions like: c_1xy^2z^3 + c_2xyz + c_3
+0x00000001

See also MultiIndexSet, FixedMultiIndexSet, Fix

source
MParT.MultiIndexSet — Type
MultiIndexSet(A::AbstractVecOrMat{<:Integer})

Create a set of MultiIndices from the rows of A.

These indices represent a basis for a multivariate expansion or, further, monotone expansion. Each element of the set is a MultiIndex representing one basis function via the degrees in each dimension.

Example

julia> # Functions like: c_1xy^2z^3 + c_2xyz + c_3
 
 julia> A = [1 2 3;1 1 1; 0 0 0];
 
-julia> mset = MultiIndexSet(A);

See also MultiIndex, FixedMultiIndexSet, Fix

source
MParT.FixedMultiIndexSet — Type
FixedMultiIndexSet(dim::Int, p::Int)

Creates a FixedMultiIndexSet with dimension dim and total order p.

A FixedMultiIndexSet is just a compressed, efficient way of representing a MultiIndexSet, but without as many bells and whistles.

See also: MultiIndex, MultiIndexSet

source
MParT.Fix — Function
Fix(mset::MultiIndexSet, compress::Bool = true)

Take mset and turn it into a FixedMultiIndexSet that can be compressed.

See also MultiIndex, MultiIndexSet, FixedMultiIndexSet

source
MParT.CreateTotalOrder — Function
CreateTotalOrder(dim::Int, p::Int)

Creates a total order p MultiIndexSet object in dimension dim.

See also: MultiIndexSet

source
MParT.Size — Function
Size(mset::MultiIndexSet)

Number of MultiIndex objects in a MultiIndexSet mset.

source
+julia> mset = MultiIndexSet(A);

See also MultiIndex, FixedMultiIndexSet, Fix

source
MParT.FixedMultiIndexSet — Type
FixedMultiIndexSet(dim::Int, p::Int)

Creates a FixedMultiIndexSet with dimension dim and total order p.

A FixedMultiIndexSet is just a compressed, efficient way of representing a MultiIndexSet, but without as many bells and whistles.

See also: MultiIndex, MultiIndexSet

source
MParT.Fix — Function
Fix(mset::MultiIndexSet, compress::Bool = true)

Take mset and turn it into a FixedMultiIndexSet that can be compressed.

See also MultiIndex, MultiIndexSet, FixedMultiIndexSet

source
MParT.CreateTotalOrder — Function
CreateTotalOrder(dim::Int, p::Int)

Creates a total order p MultiIndexSet object in dimension dim.

See also: MultiIndexSet

source
MParT.Size — Function
Size(mset::MultiIndexSet)

Number of MultiIndex objects in a MultiIndexSet mset.

source
diff --git a/dev/trainmap/index.html b/dev/trainmap/index.html index 05d64a0..b126271 100644 --- a/dev/trainmap/index.html +++ b/dev/trainmap/index.html @@ -18,7 +18,7 @@ julia> obj3 = CreateGaussianKLObjective(train, test); -julia> obj4 = CreateGaussianKLObjective(train, test, outDim);

See also TrainMap, TrainOptions

source
MParT.TrainOptions — Function
`TrainOptions(;kwargs...)`

Creates options for using TrainMap to train a transport map.

See example for possible arguments.

Examples

julia> TrainOptions(opt_alg="LD_SLSQP", opt_maxeval = 1_000_000)
+julia> obj4 = CreateGaussianKLObjective(train, test, outDim);

See also TrainMap, TrainOptions

source
MParT.TrainOptions — Function
`TrainOptions(;kwargs...)`

Creates options for using TrainMap to train a transport map.

See example for possible arguments.

Examples

julia> TrainOptions(opt_alg="LD_SLSQP", opt_maxeval = 1_000_000)
 opt_alg = LD_SLSQP
 opt_stopval = -inf
 opt_ftol_rel = 0.001
@@ -28,4 +28,4 @@
 opt_maxeval = 1000000
 opt_maxtime = inf
 verbose = 0
-

See also TrainMap, CreateGaussianKLObjective

source
MParT.TrainMap — Function
TrainMap(map, obj::MapObjective, opts::TrainOptions)

Trains map according to the objective obj with training options opts.

source
MParT.TestError — Function
TestError(obj::MapObjective, map)

Uses the test dataset in obj to evaluate the error of map.

source
+

See also TrainMap, CreateGaussianKLObjective

source
MParT.TrainMap — Function
TrainMap(map, obj::MapObjective, opts::TrainOptions)

Trains map according to the objective obj with training options opts.

source
MParT.TestError — Function
TestError(obj::MapObjective, map)

Uses the test dataset in obj to evaluate the error of map.

source