agentMet4FoF use case anomaly detection based on machine-learning

This is supported by European Metrology Programme for Innovation and Research (EMPIR) under the project Metrology for the Factory of the Future (Met4FoF), project number 17IND12.

Anomaly detection

With the provided code we showcase an agent-based machine learning approach for online anomaly detection of (in our case simulated) sensor readings.

Getting started

In case you are using PyCharm, you will already find proper run configurations at the appropriate place in the IDE. It expects that you have prepared and defined a default interpreter.

If you are not using PyCharm, of course you can run the script files as usual.

If you have any questions please get in touch with the author.

Dependencies

To install all dependencies in virtual environment based on Python version 3.7 first install pip-tools and afterwards use our prepared requirements.txt to get everything ready.

Create a virtual environment on Windows

In your Windows command prompt execute the following to set up a virtual environment in a folder of your choice.

> python -m venv my_anomaly_detection_use_case_env
> my_anomaly_detection_use_case_env\Scripts\activate.bat
(my_anomaly_detection_use_case_env) > pip install --upgrade pip setuptools pip-tools
(my_anomaly_detection_use_case_env) > pip-sync

Create a virtual environment on Mac and Linux

In your terminal execute the following to set up a virtual environment in a folder of your choice.

$ python3.7 -m venv my_anomaly_detection_use_case_env
$ source my_anomaly_detection_use_case_env/bin/activate
(my_anomaly_detection_use_case_env) $ pip install --upgrade pip setuptools pip-tools
(my_anomaly_detection_use_case_env) $ pip-sync

Scripts

The interesting parts you find in the file

• agentMET4FOF_anomaly_detection/anomaly_detection.py

Orphaned processes

In the event of agents not terminating cleanly, you can end all Python processes running on your system (caution: the following commands affect all running Python processes, not just those that emerged from the agents).

Killing all Python processes in Windows

In your Windows command prompt execute the following to terminate all python processes.

> taskkill /f /im python.exe /t
>

Killing all Python processes on Mac and Linux

In your terminal execute the following to terminate all python processes.

$ pkill python
$

References

For details about the agents refer to the upstream repository agentMET4FOF

Screenshot of web visualization

Developing

For development and testing you should as well install the development dependencies provided in the dev-requirements.txt for Python 3.8 as well.

$ source my_anomaly_detection_use_case_env/bin/activate
(my_anomaly_detection_use_case_env) $ pip-sync dev-requirements.txt requirements.txt

You will find another run configuration for the test suite in your PyCharm IDE.