torch_openreml.covariance.SimpleMatrix¶

class torch_openreml.covariance.SimpleMatrix(n, call, manual_grad=None, default=0.0)[source]¶

Bases: Matrix

A covariance matrix for simple, function-based parameterisations.

This is the easiest way to use REML with a custom covariance structure: provide the number of parameters and a function that maps a flat parameter tensor to the covariance matrix. All parameters are free and use an identity transform (unconstrained). The default argument sets the value used for each free parameter when none are provided.

For more advanced needs (custom transforms, fixed parameters, manual gradients), subclass Matrix directly.

Parameters:

n (int) – Number of free parameters.
call (callable) – Function with signature call(free_params) -> torch.Tensor that constructs the covariance matrix from a flat 1D parameter tensor.
manual_grad (callable, optional) – Function with signature manual_grad(free_params) -> (grad, grad_names) for a closed-form Jacobian. If None (default), automatic differentiation is used.
default (float or torch.Tensor, optional) – Default value for each parameter. Passed to simple_param_specs(). Defaults to 0.0.

Example:

import torch
from torch_openreml.covariance import SimpleMatrix

def my_v(free_params):
    n = free_params.shape[0]
    return torch.diag(free_params)

mat = SimpleMatrix(n=3, call=my_v)
mat(torch.tensor([1.0, 2.0, 3.0]))

tensor([[1., 0., 0.],
        [0., 2., 0.],
        [0., 0., 3.]])

mat.grad(torch.tensor([1.0, 2.0, 3.0]))

(tensor([[[1., 0., 0.],
          [0., 0., 0.],
          [0., 0., 0.]],
 
         [[0., 0., 0.],
          [0., 1., 0.],
          [0., 0., 0.]],
 
         [[0., 0., 0.],
          [0., 0., 0.],
          [0., 0., 1.]]]),
 ['theta_0', 'theta_1', 'theta_2'])

Initialize a covariance matrix with parameter specifications.

Parameters:

shape (tuple or None) – Expected output dimensions of the constructed matrix. Used for validation; the actual shape may be set by subclasses.
param_specs (dict) – Parameter specifications. Keys should be strings representing parameter names. Values should be dictionaries containing the specification for each parameter. Each specification dictionary should contain the keys "fixed", "default", and "trans", representing whether the parameter is fixed or free (bool), the default value (1D torch.Tensor), and the transform (Transform), respectively.

Raises:

TypeError – If param_specs does not follow any of the requirements listed in the argument description, or if shape is not a tuple or torch.Size.
ValueError – If shape values are non-negative.

Methods

`__call__`([free_params])	Construct the matrix from a flat parameter tensor.
`auto_grad`([free_params])	Compute the Jacobian of `build()` with respect to free parameters using automatic differentiation.
`build_params`([free_params, include_fixed, ...])	Construct the full parameter tensor from free parameters.
`get_intermediates`(params)	Retrieve cached intermediate computation results if still valid.
`grad`([free_params])	Compute the Jacobian of `__call__()` with respect to trainable parameters.
`manual_grad`([free_params])	Compute the Jacobian of `__call__()` with respect to free parameters using a closed-form analytic expression.
`map_theta_to_dv`(theta)	An interface compatible with `torch_openreml.REML` that maps parameters to the matrix Jacobian.
`map_theta_to_v`(theta)	An interface compatible with `torch_openreml.REML` that maps parameters to a matrix.
`reset_intermediates`()	Clear the intermediate computation cache.
`set_intermediates`(params, intermediates)	Cache intermediate computation results keyed by parameter hash.
`trans_grad`([free_params])	Compute the element-wise derivative of the free parameter transforms.

Attributes

`fixed_param_defaults`	Fixed parameter defaults.
`fixed_param_index`	Index of fixed parameters.
`fixed_param_names`	Fixed parameter names.
`fixed_param_trans`	Transforms for fixed parameters.
`free_param_defaults`	Free parameter defaults.
`free_param_index`	Index of free parameters.
`free_param_names`	Free parameter names.
`free_param_trans`	Transforms for free parameters.
`num_fixed_params`	Total number of fixed parameters.
`num_free_params`	Total number of free parameters.
`num_params`	Total number of parameters.
`param_defaults`	Parameter defaults.
`param_names`	Parameter names.
`param_specs`	Parameter specifications.
`param_trans`	Parameter transforms.
`repr_dict`	Key-value pairs used to build the string representation.
`shape`	Output matrix shape.

__call__(free_params=None)[source]¶

Construct the matrix from a flat parameter tensor.

Must be implemented by subclasses. Implementations should convert free_params via build_params() to validate, include fixed parameters, and apply transforms before any computation.

Parameters:: free_params (torch.Tensor or dict) – Flat 1D parameter tensor or parameter dictionary. If omitted, default values are used. Default: None.
Returns:: Constructed matrix of shape shape.
Return type:: torch.Tensor

manual_grad(free_params=None)[source]¶

Compute the Jacobian of __call__() with respect to free parameters using a closed-form analytic expression.

This method is optional. When implemented by a subclass, grad() will invoke it in preference to auto_grad() under the default grad mode. If not implemented, calling this method raises NotImplementedError and grad() falls back to automatic differentiation.

Implementations must satisfy the following contract:

Return (None, []) if all parameters are fixed.
Return a 3D gradient tensor of shape (num_free_params, *shape) and a matching list of parameter names.
Apply transform derivatives from trans_grad() via the chain rule so that gradients are with respect to the raw (untransformed) parameters.

Parameters:: free_params (torch.Tensor or dict) – Flat 1D parameter tensor or parameter dictionary. If omitted, default values are used. Default: None.
Returns:: (grad, grad_names), where grad is a 3D tensor of shape (num_free_params, *shape) and grad_names is a list of the corresponding parameter names. Returns (None, []) if all parameters are fixed.
Return type:: tuple
Raises:: NotImplementedError – If the subclass does not provide an analytic gradient. grad() catches this and falls back to auto_grad().