yurita

Leverages Apache Spark for distributed processing of large datasets, making it production-ready for big data workflows as emphasized in the README's focus on large-scale models.

Uses a builder pattern for configurable statistical methods like categorical averaging and entropy, allowing data scientists to compose custom anomaly detection workflows, as shown in the sample code.

Supports fixed and sliding window definitions for time-series analysis, which is critical for monitoring and security applications dealing with streaming data.

Handles both categorical and numerical columns with tailored functions, enhancing versatility in detecting anomalies across diverse data types as highlighted in the key features.