This folder contains training and inference scripts for two models based on different technologies: MLP (multi-layer perceptron) using scikit-learn and an LSTM-based neural network using TensorFlow. Each model has its own folder with training and testing scripts. Developers interested in DGA detection in nDPI should also visit this folder containing the original ML implementation.

The test scripts only show how to use an already-trained model.

To install the necessary dependencies, run

pip install -r requirements.txt

Folder: scikit-learn_tests.

To train the MLP model, run the training script:

python scikit-learn_tests/training_script.py

After training, you can perform inference using the test script:

python scikit-learn_tests/test_script.py

Folder: tensorflow_tests.

To train the LSTM model, run the training script

python tensorflow_tests/training_script.py

Once training is complete, you can run inference on the test set with

python tensorflow_tests/test_script.py