Dataset from a NumPy array

In this example, we will see how to build a Dataset from an NumPy array.

from __future__ import annotations

from numpy import concatenate
from numpy.random import default_rng

from gemseo import configure_logger
from gemseo.datasets.dataset import Dataset

configure_logger()

rng = default_rng(1)

Let us consider three parameters \(x_1\), \(x_2\) and \(x_3\) of size 1, 2 and 3 respectively. We generate 5 random samples of the inputs where

• x_1 is stored in the first column,

• x_2 is stored in the 2nd and 3rd columns

and 5 random samples of the outputs:

n_samples = 5
inputs = rng.random((n_samples, 3))
outputs = rng.random((n_samples, 3))
data = concatenate((inputs, outputs), 1)

A dataset with default names

We create a Dataset from the NumPy array only and let GEMSEO give default names to its columns:

dataset = Dataset.from_array(data)
dataset

GROUP	parameters
VARIABLE	x_0	x_1	x_2	x_3	x_4	x_5
COMPONENT	0	0	0	0	0	0
0	0.511822	0.950464	0.144160	0.453498	0.134042	0.403113
1	0.948649	0.311831	0.423326	0.203455	0.262313	0.750365
2	0.827703	0.409199	0.549594	0.280409	0.485191	0.980737
3	0.027559	0.753513	0.538143	0.961657	0.724790	0.541227
4	0.329732	0.788429	0.303195	0.276891	0.160652	0.969925

A dataset with custom names

We can also pass the names and sizes of the variables:

names_to_sizes = {"x_1": 1, "x_2": 2, "y_1": 3}
dataset = Dataset.from_array(data, ["x_1", "x_2", "y_1"], names_to_sizes)
dataset

GROUP	parameters
VARIABLE	x_1	x_2		y_1
COMPONENT	0	0	1	0	1	2
0	0.511822	0.950464	0.144160	0.453498	0.134042	0.403113
1	0.948649	0.311831	0.423326	0.203455	0.262313	0.750365
2	0.827703	0.409199	0.549594	0.280409	0.485191	0.980737
3	0.027559	0.753513	0.538143	0.961657	0.724790	0.541227
4	0.329732	0.788429	0.303195	0.276891	0.160652	0.969925

Warning

The number of variables names must be equal to the number of columns of the data array. Otherwise, the user has to specify the sizes of the different variables by means of a dictionary and be careful that the total size is equal to this number of columns.

A dataset with custom groups

We can also use the notions of groups of variables:

groups = {"x_1": "inputs", "x_2": "inputs", "y_1": "outputs"}
dataset = Dataset.from_array(data, ["x_1", "x_2", "y_1"], names_to_sizes, groups)
dataset

GROUP	inputs			outputs
VARIABLE	x_1	x_2		y_1
COMPONENT	0	0	1	0	1	2
0	0.511822	0.950464	0.144160	0.453498	0.134042	0.403113
1	0.948649	0.311831	0.423326	0.203455	0.262313	0.750365
2	0.827703	0.409199	0.549594	0.280409	0.485191	0.980737
3	0.027559	0.753513	0.538143	0.961657	0.724790	0.541227
4	0.329732	0.788429	0.303195	0.276891	0.160652	0.969925

Note

The groups are specified by means of a dictionary where indices are the variables names and values are the groups. If a variable is missing, the default group Dataset.DEFAULT_GROUP is considered.