This project is about preprocessing log lines in order to allow efficient and effective data analysis.
Please install the logdata-anomaly-miner (AMiner) first.
Next you can simply clone the repository from git:
git clone https://github.com/ait-aecid/log-preprocessor
A small example for how to use the LogData class from LogData.py is given in example.ipynb. The LogData class solely requires the parser name and the path to the folder containing the data as input parameters.
The file AMinerModel.py additionally contains the AMinerModel, which allows easy usage of the AMiner in a 'scikit-learn-like' way. Training and testing the model is thereby already possible with a few lines of code. Be aware that the execution in Jupyter notebooks requires sudo priviledges and will therefore ask for your password (this is not a good practice in general but necessary since the AMiner requires sudo priviledges).