Gaussian Mixtures

Load Iris dataset and create clusters.

from __future__ import annotations

from numpy import array

from gemseo import configure_logger
from gemseo import create_benchmark_dataset
from gemseo.datasets.dataset import Dataset
from gemseo.mlearning import create_clustering_model
from gemseo.post.dataset.scatter_plot_matrix import ScatterMatrix

configure_logger()
<RootLogger root (INFO)>

Create dataset

We import the Iris benchmark dataset through the API.

iris = create_benchmark_dataset("IrisDataset")

Extract inputs as a new dataset

data = iris.get_view(group_names=iris.PARAMETER_GROUP).to_numpy()
variables = iris.get_variable_names(iris.PARAMETER_GROUP)
variables
['petal_length', 'petal_width', 'sepal_length', 'sepal_width']
dataset = Dataset.from_array(data, variables)

Create clustering model

We know that there are three classes of Iris plants. We will thus try to identify three clusters.

model = create_clustering_model("GaussianMixture", data=dataset, n_components=3)
model.learn()
model
GaussianMixture(n_components=3, random_state=0, var_names=None)
  • based on the scikit-learn library
  • built from 150 learning samples


Predict output

Once it is built, we can use it for prediction.

input_value = {
    "sepal_length": array([4.5]),
    "sepal_width": array([3.0]),
    "petal_length": array([1.0]),
    "petal_width": array([0.2]),
}
output_value = model.predict(input_value)
output_value
2

Plot clusters

Show cluster labels

dataset.add_variable("gm_specy", model.labels.reshape((-1, 1)), "labels")
ScatterMatrix(dataset, kde=True, classifier="gm_specy").execute(save=False, show=True)
plot gaussian mixture
[<Figure size 640x480 with 16 Axes>]

Total running time of the script: (0 minutes 0.761 seconds)

Gallery generated by Sphinx-Gallery