IndependentMCSampler

class axtreme.sampling.independent_sampler.IndependentMCSampler(sample_shape: Size, seed: int | None = None, **kwargs: Any)

Bases: MCSampler, ABC

Abstract base class for MCSamplers that independantly apply the same set of base samples at each x point.

This should be used when you want to ignore the covariance between different x points, and sample the X output space independently.

Note

Dimensions of the posterior are described using botorch notation (detailed in glassary.md) where:

• *b: batch shape (can have arbitrary dimensionality).

• n: the number of x input points.

• m: dimensionality of targets space.

For example:

• MultivariateNormal: (*b, n)

• MultitaskMultivariateNormal: (*b, n, m)

Dev Notes:

• Is the shape of the base samples up to us to define, or are there some other parts of the system that expect MCSampler to have certain shape.

  Answer: MCSampler does not define base_sample - we can do what we want.

• MultivariateNormal and MultitaskMultivariateNormal both requires shape (*sample_shape, *posterior.shape() for base_samples in posterior.rsample_from_base_samples

  • For MultitaskMultivariateNormal: posterior.shape() = (*b,n,m)

  • For MultivariateNormal: posterior.shape() = (*b,n)

Design decision: we decide to make m explicit throughout.

• e.g base samples will always have shape (*sample_shape,m)

• This makes our code cleaner. When then convert back to implicit m dimension when calling posterior.rsample_from_base_samples

__init__(sample_shape: Size, seed: int | None = None, **kwargs: Any) → None

Abstract base class for samplers.

Parameters:

• sample_shape – The sample_shape of the samples to generate. The full shape of the samples is given by posterior._extended_shape(sample_shape).

• seed – An optional seed to use for sampling.

• **kwargs – Catch-all for deprecated kwargs.

Methods

__init__(sample_shape[, seed])	Abstract base class for samplers.
add_module(name, module)	Add a child module to the current module.
apply(fn)	Apply fn recursively to every submodule (as returned by .children()) as well as self.
bfloat16()	Casts all floating point parameters and buffers to bfloat16 datatype.
buffers([recurse])	Return an iterator over module buffers.
children()	Return an iterator over immediate children modules.
compile(*args, **kwargs)	Compile this Module's forward using torch.compile().
cpu()	Move all model parameters and buffers to the CPU.
cuda([device])	Move all model parameters and buffers to the GPU.
double()	Casts all floating point parameters and buffers to double datatype.
eval()	Set the module in evaluation mode.
extra_repr()	Set the extra representation of the module.
float()	Casts all floating point parameters and buffers to float datatype.
forward(posterior)	Draw samples from the posterior, treating each of the n input points as independant.
get_buffer(target)	Return the buffer given by target if it exists, otherwise throw an error.
get_extra_state()	Return any extra state to include in the module's state_dict.
get_parameter(target)	Return the parameter given by target if it exists, otherwise throw an error.
get_submodule(target)	Return the submodule given by target if it exists, otherwise throw an error.
half()	Casts all floating point parameters and buffers to half datatype.
ipu([device])	Move all model parameters and buffers to the IPU.
load_state_dict(state_dict[, strict, assign])	Copy parameters and buffers from state_dict into this module and its descendants.
modules()	Return an iterator over all modules in the network.
named_buffers([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters([recurse])	Return an iterator over module parameters.
register_backward_hook(hook)	Register a backward hook on the module.
register_buffer(name, tensor[, persistent])	Add a buffer to the module.
register_forward_hook(hook, *[, prepend, ...])	Register a forward hook on the module.
register_forward_pre_hook(hook, *[, ...])	Register a forward pre-hook on the module.
register_full_backward_hook(hook[, prepend])	Register a backward hook on the module.
register_full_backward_pre_hook(hook[, prepend])	Register a backward pre-hook on the module.
register_load_state_dict_post_hook(hook)	Register a post hook to be run after module's load_state_dict is called.
register_module(name, module)	Alias for add_module().
register_parameter(name, param)	Add a parameter to the module.
register_state_dict_pre_hook(hook)	Register a pre-hook for the state_dict() method.
requires_grad_([requires_grad])	Change if autograd should record operations on parameters in this module.
set_extra_state(state)	Set extra state contained in the loaded state_dict.
share_memory()	See torch.Tensor.share_memory_().
state_dict(*args[, destination, prefix, ...])	Return a dictionary containing references to the whole state of the module.
to(*args, **kwargs)	Move and/or cast the parameters and buffers.
to_empty(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
train([mode])	Set the module in training mode.
type(dst_type)	Casts all parameters and buffers to dst_type.
xpu([device])	Move all model parameters and buffers to the XPU.
zero_grad([set_to_none])	Reset gradients of all model parameters.

Attributes

T_destination
call_super_init
dump_patches
training

forward(posterior: Posterior) → Tensor

Draw samples from the posterior, treating each of the n input points as independant.

Parameters:

posterior – The posterior which should be sampled.

Returns:

(*sample_shape, *b, n, m) where: - *b: is the posterior batch shape - n: number of input points in posterior - m: dimensionality of target

NOTE: While MultitaskMultivariateNormal and MultivariateNormal have different shapes ((*b,n,m) and (*b,n)) the output of this function is always of shape (*sample_shape, *b, n, m). This is consistent with the behaviour of MCSampler.

Return type:

Returns a samples of the posterior with shape