ramsey.data
Functionality for loading and sampling data sets.
ramsey.data.m4_data(interval: str = 'hourly', drop_na: bool = True)
Load a data set from the M4 competition.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
interval |
str
|
either of "hourly", "daily", "weekly", "monthly", "yearly" |
'hourly'
|
drop_na |
bool
|
drop rows that contain NA values |
True
|
Returns:
Type | Description |
---|---|
NamedTuple
|
returns a named tuple with outputs (y), inputs (x), and training and testing indexes for the input-output paris |
ramsey.data.sample_from_gaussian_process(rng_key, batch_size=10, num_observations=100, rho=None, sigma=None)
Sample from a Gaussian process.
Creates samples from a Gaussian process with exponentiated quadratic
covariance function. For each batch, chooses a new hyperparameter
configuration where rho
, the kernel lengthscale is drawn
from an InverseGamma(1, 1)
and sigma, the kernel lengthscale, is drawn
from an InverseGamma(5, 5)
.
The inputs, x
of the Gaussian process have dimensionality
:math:b \times n \times 1
, where b
is the batch size and n
is the
number of observations per batch. The outputs and latent functions
realizations have dimension :math:b \times n \times 1
as well.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
rng_key |
a random key for seeding |
required | |
batch_size |
size of batch |
10
|
|
num_observations |
number of observations per batch |
100
|
|
rho |
the lengthscale of the kernel function |
None
|
|
sigma |
the standard deviation of the kernel function |
None
|
Returns:
Type | Description |
---|---|
NamedTuple
|
a tuple consisting of outputs (y), inputs (x) and latent GP realization (f) where |
ramsey.data.sample_from_sine_function(rng_key, batch_size=10, num_observations=100)
Sample from a noisy sine function.
Creates samples from a noisy sine functions. For each batch, chooses a new hyperparameters configuration.
The inputs, x
of the sine function have dimensionality
:math:b \times n \times 1
, where b
is the batch size and n
is the
number of observations per batch. The outputs and latent functions
realizations have dimension :math:b \times n \times 1
as well.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
rng_key |
a random key for seeding |
required | |
batch_size |
size of batch |
10
|
|
num_observations |
number of observations per batch |
100
|
|
rho |
the lengthscale of the kernel function |
required | |
sigma |
the standard deviation of the kernel function |
required |
Returns:
Type | Description |
---|---|
NamedTuple
|
a tuple consisting of outputs (y), inputs (x) and latent GP realization (f) where |