# Copyright 2018 Amazon.com, Inc. or its affiliates. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License").
# You may not use this file except in compliance with the License.
# A copy of the License is located at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# or in the "license" file accompanying this file. This file is distributed
# on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either
# express or implied. See the License for the specific language governing
# permissions and limitations under the License.
import math
from typing import Dict, Tuple
from gluonts.mx import Tensor
from gluonts.mx.distribution.distribution import getF, softplus
from . import Kernel, KernelOutputDict
[docs]class RBFKernel(Kernel):
r"""
Computes a covariance matrix based on the RBF (squared exponential) kernel
between inputs :math:`\mathbf{x_1}` and :math:`\mathbf{x_2}`:
:math:`k_{\text{RBF}}(\mathbf{x_1}, \mathbf{x_2}) = \theta_0 \exp \left
( -\frac{\|\mathbf{x_1} - \mathbf{x_2}\|^2}
{2\theta_1^2} \right)`,
where :math:`\theta_0` is the amplitude parameter and
:math:`\theta_1` is the length scale parameter.
"""
# noinspection PyMethodOverriding,PyPep8Naming
def __init__(
self, amplitude: Tensor, length_scale: Tensor, F=None
) -> None:
"""
Parameters
----------
amplitude : Tensor
RBF kernel amplitude hyper-parameter of shape (batch_size, 1, 1).
length_scale : Tensor
RBF kernel length scale hyper-parameter of of shape
(batch_size, 1, 1).
F : ModuleType
A module that can either refer to the Symbol API or the NDArray
API in MXNet.
"""
self.F = F if F else getF(amplitude)
self.amplitude = amplitude
self.length_scale = length_scale
# noinspection PyMethodOverriding,PyPep8Naming
[docs] def kernel_matrix(self, x1: Tensor, x2: Tensor) -> Tensor:
"""
Parameters
--------------------
x1 : Tensor
Feature data of shape (batch_size, history_length, num_features).
x2 : Tensor
Feature data of shape (batch_size, history_length, num_features).
Returns
--------------------
Tensor
RBF kernel matrix of shape (batch_size, history_length,
history_length).
"""
self._compute_square_dist(self.F, x1, x2)
return self.F.broadcast_mul(
self.amplitude,
self.F.exp(
self.F.broadcast_div(
-self.square_dist, 2 * self.length_scale**2
)
),
)
[docs]class RBFKernelOutput(KernelOutputDict):
args_dim: Dict[str, int] = {"amplitude": 1, "length_scale": 1}
kernel_cls: type = RBFKernel
# noinspection PyMethodOverriding,PyPep8Naming
[docs] def gp_params_scaling(
self, F, past_target: Tensor, past_time_feat: Tensor
) -> Tuple[Tensor, Tensor, Tensor]:
"""
This function returns the scales for the GP RBF Kernel hyper-parameters
by using the standard deviations of the past_target and
past_time_features.
Parameters
----------
F
A module that can either refer to the Symbol API or the NDArray
API in MXNet.
past_target
Training time series values of shape (batch_size, context_length).
past_time_feat
Training features of shape (batch_size, context_length,
num_features).
Returns
-------
Tuple
Two scaled GP hyper-parameters for the RBF Kernel and scaled model
noise hyper-parameter. Each is a Tensor of shape
(batch_size, 1, 1).
"""
axis = 1
sigma_scaling = (
self.compute_std(F, past_target, axis=axis) / math.sqrt(2)
).expand_dims(axis=axis)
amplitude_scaling = sigma_scaling**2
length_scale_scaling = F.broadcast_mul(
F.mean(self.compute_std(F, past_time_feat, axis=axis)),
F.ones_like(amplitude_scaling),
)
return amplitude_scaling, length_scale_scaling, sigma_scaling
# noinspection PyMethodOverriding,PyPep8Naming
[docs] def domain_map(self, F, amplitude, length_scale):
"""
This function applies the softmax to the RBF Kernel hyper-parameters.
Parameters
----------
F
A module that can either refer to the Symbol API or the NDArray
API in MXNet.
amplitude
RBF kernel amplitude hyper-parameter of shape (batch_size, 1, 1).
length_scale
RBF kernel length scale hyper-parameter of of shape
(batch_size, 1, 1).
Returns
-------
Tuple[Tensor, Tenspr]
Two GP RBF kernel hyper-parameters.
Each is a Tensor of shape: (batch_size, 1, 1).
"""
amplitude = softplus(F, amplitude)
length_scale = softplus(F, length_scale)
return amplitude, length_scale