SpecMix is a Python package that has implementations of SpecMix and onlyCat, clustering algorithms that rely on spectral graph theory for mixed-type and categorical data.
Both SpecMix and onlyCat are based off of the clustering algorithms in scikit-learn.
from SpecMix.specmix import SpecMix
from examples.synthetic_dataset_generation import generate_mixed_dataset
#Generate a synthetic dataset with 2 numerical features, 2 categorical features, 3 clusters, 0.3 noise
df = generate_mixed_dataset(n_samples=1000, n_numerical_features=3, n_categorical_features=2, n_clusters=3, p=0.3)
#Initialize the SpecMix algorithm with 3 clusters
specmix = SpecMix(n_clusters=3, random_state=0)
#Fit the algorithm to the dataset
specmix.fit(df)
#Print the cluster labels
print(specmix.labels_)This example shows the use of SpecMix on a synthetic dataset with 2 numerical features, 2 categorical features, 3 clusters, and 0.3 noise. The algorithm is initialized with 3 clusters and then fit to the dataset. The cluster labels are then printed. For more detailed usage instructions and examples, please refer to the demo.ipynb notebook included in this repository.
If you encounter any bugs or issues, feel free to open an issue on the GitHub repository. Please include a detailed description of the bug or issue, as well as any relevant code snippets or error messages.