PEAK is a Python tool designed to make easier the basic steps of pattern recognition, data collection, data exploration, data correlation, regression analysis, and/or classification. Therefore, PEAK allows less experienced users to reduce the time required for analysing data and promote the dis-covery of unknown relationships between different data.
wget https://github.com/Pex2892/PEAK/archive/main.zip
unzip main.zipor
git clone https://github.com/Pex2892/PEAK.gitA typical user can install the libraries using the following command:
python3 -m pip install -r requirements.txtBefore starting the analysis, PEAK requires to configure "settings.cfg". In this file, there are four sections.
- Settings: this section contains the general settings of the tool, such as the seed, the number of CPUs, and the clearing of the previous results.
- Dataset: this section contains parameters related to the dataset, such as file name, separator, and the number of rows to skip.
- Regression: it contains all the settings needed to start a regression analysis, such as the dependent variable name and the parameters useful for the resampling phase.
- Classification: it contains all the settings required for classification, such as the variable to be classified, the targets and the parameters needed for the resampling phase.
After that, you can run PEAK analysis:
python3 main.py