Metadata-Version: 2.1
Name: anomaly-detection-models
Version: 0.1
Summary: Models for anomaly detection; see e.g. https://arxiv.org/abs/2009.02205
Home-page: https://github.com/luclepot/anomaly_detection_models
Author: Luc Le Pottier
Author-email: luclepot@umich.edu
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: matplotlib
Requires-Dist: numpy
Requires-Dist: Keras (>=2.2.4)
Requires-Dist: scikit-learn (>=0.21.2)
Requires-Dist: pickleshare (>=0.7.5)
Requires-Dist: pandas (>=0.25.1)

# anomaly_detection_models

Repository with some useful anomaly detection model definitions.

## install

clone this repository, and run 

    pip install . [--user]

with the ``--user`` argument specifying local installation.

## usage

import models directly or subclass `anomaly_detection_base` to make a new model (instructions in-source)

## example

see [`demos/test.ipynb`](/demos/test.ipynb) for an example. general usage is like sklearn, as

```
from anomaly_detection_models import SACWoLa

sacwola = SACWoLa(epochs=10, lambda_=1.2)
sacwola.fit(x, y_sim, y_sb)

pred = sacwola.predict(x_test)
```

