microcluster_benchmark.py

import torch
import time
import numpy as np
from sklearn.datasets import make_blobs
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from microcluster_module import MClassification

def generate_data(n_samples=1000, n_features=2, centers=5):
    X, y = make_blobs(n_samples=n_samples, n_features=n_features, centers=centers, random_state=42)
    return X, y

def main():
    # Generate synthetic data
    X, y = generate_data()
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, random_state=42)

    # Convert to PyTorch tensors
    X_train_tensor = torch.tensor(X_train, dtype=torch.float32)
    y_train_tensor = torch.tensor(y_train, dtype=torch.int)
    X_test_tensor = torch.tensor(X_test, dtype=torch.float32)
    y_test_tensor = torch.tensor(y_test, dtype=torch.int)

    # Initialize and train model
    model = MClassification(max_radius=0.2, max_clusters=100)
    model.fit_initial(X_train_tensor, y_train_tensor)

    # Benchmarking on test data
    start_time = time.time()
    predictions = [model.predict_and_update(x) for x in X_test_tensor]
    end_time = time.time()

    # Calculate accuracy
    accuracy = accuracy_score(y_test_tensor.numpy(), predictions)

    # Report results
    execution_time = end_time - start_time
    print(f'Accuracy: {accuracy*100:.2f}%')
    print(f'Execution Time: {execution_time:.2f} seconds')

if __name__ == "__main__":
    main()