Metadata-Version: 2.1
Name: streamad
Version: 0.1.0
Summary: Python toolbox for stream anomaly (outlier) detection.
Home-page: https://github.com/Fengrui-Liu/StreamAD
Author: liufr
Author-email: liufengrui18z@ict.ac.cn
License: UNKNOWN
Platform: all
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Education
Classifier: Intended Audience :: Financial and Insurance Industry
Classifier: Intended Audience :: Science/Research
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Information Technology
Classifier: Programming Language :: Python :: 3.5
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: License :: OSI Approved :: BSD License
Requires-Dist: numpy (>=1.13)
Requires-Dist: pandas (>=0.25)
Requires-Dist: scikit-learn (>=0.19.1)
Requires-Dist: scipy (>=0.19.1)

.. image:: ./docs/source/images/logo_htmlwithname.svg
    :align: center

An anomaly detection package for streaming data.

`Documentation <https://www.liufr.com/StreamAD/>`_


------------------------------------------------------

Why StreamAD
=============


Purpose & Advantages
^^^^^^^^^^^^^^^^^^^^^^^^^^^

StreamAD focuses on streaming settings, where data features evolve and distributions change over time. To prevent the failure of static models, StreamAD can correct its model as needed.

Incremental & Continual
^^^^^^^^^^^^^^^^^^^^^^^^^^^

StreamAD loads static datasets to a stream generator and feed a single observation at a time to any model in StreamAD. Therefore it can be used to simulate real-time applications and process streaming data.


Models & Algorithms
^^^^^^^^^^^^^^^^^^^^^^^^^^^

StreamAD collects open source implementations and reproduce state-of-the-art papers. Thus, it can also be used as an benchmark for academic.


Efficient & Scalability:
^^^^^^^^^^^^^^^^^^^^^^^^^^^

StreamAD concerns about the running time, resources usage and usability of different models. It is implemented by python and you can design your own algorithms and run with StreamAD.



Free & Open Source Software (FOSS)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

`StreamAD` is distributed under `BSD License 3.0 <https://github.com/Fengrui-Liu/StreamAD/master/LICENSE>`_ and favors FOSS principles.


------------------------------------------------------

Installation
============


The StreamAD framework can be installed via:


.. code-block:: bash

    pip install -U StreamAD


Alternatively, you can install the library directly using the source code in Github repository by:


.. code-block:: bash

    git clone https://github.com/Fengrui-Liu/StreamAD.git
    cd StreamAD
    pip install .

------------------------------------------------------

Models
===================


* `KNN CAD <https://arxiv.org/abs/1608.04585>`_
* `xStream <https://cmuxstream.github.io/>`_
* `SPOT <https://dl.acm.org/doi/10.1145/3097983.3098144>`_
* LSTMAutoencoder

