DeepOD
Visit ToolDeepOD is an open-source Python library for deep learning-based outlier and anomaly detection. It supports both tabular and time-series data, offering 27 algorithms and a comprehensive testbed for research.
At a glance
Trending
DeepOD is an open-source Python library for deep learning-based outlier and anomaly detection. It supports both tabular and time-series data, offering 27 algorithms and a comprehensive testbed for research.
Trending
About
DeepOD is an open-source Python library designed for deep learning-based outlier and anomaly detection. It provides a unified API across 27 different algorithms, supporting both tabular and time-series data types. The library features state-of-the-art models including reconstruction-, representation-learning-, and self-supervised-based deep learning methods. DeepOD also includes a comprehensive testbed, highly recommended for academic research, which allows direct testing of various models on benchmark datasets. Future updates plan to support additional data types like images, graphs, logs, and traces. Users can also plug in diverse network structures such as LSTM, GRU, TCN, Conv, and Transformer for time-series data.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending