Installation

To install the library, follow these steps.

• Run addpaths.m.

• Enjoy.

Note

The data generated by the library are by default stored into a folder data, contained in the main folder of the library. If you want to modify this location, just copy in the main folder the file options_example.ini into options.ini (not git tracked) and customize the field datapath.

Note

A collection of pre-trained ANNs is available in the repo model-learning_data. To use them, clone or download the repo and make the datapath field point to the path of the repo (or alternatively copy the content of the repo into the datapath folder).

Python wrapper

The library provides a Python interface to deploy the trained model. The module can be loaded as:

import pyModelLearning

To install it, you can choose among the following options.

Option 1: Install the library by setuptools

From the main folder of this repository run:

$ pip install . --use-feature=in-tree-build

Option 2: Add the library path to Python paths

Linux / macOS

Add to your .bashrc:

export PYTHONPATH="${PYTHONPATH}:/path/to/model-learning"

Windows

Add the path of model-learning to the PYTHONPATH environment variable. Alternatively, you can write the path of model-learning inside a pth file, as described here.

Option 3: Add the library path within your python script

Put the following lines at the beginning of your Python script:

import sys
sys.path.append("/path/to/model-learning")
import pyModelLearning