This article explains why Convolutional Neural Networks (CNN) are useful for function identification in reverse engineering, and provides a Python implementation using PyTorch to identify the start of functions in binaries.
Scikit-learn — the go-to library for machine learning offering a user friendly, consistent interface.
Pycaret — lowering the entry point for machine learning with low code, automated and end to end solutions.
PyTorch — build and deploy powerful, scalable neural networks with its highly flexible architecture.
TensorFlow — one of the most mature deep learning libraries, highly flexible and suited to a wide range of applications.
Keras — TensorFlow made simple.
FastAI — makes deep learning more accessible with a high-level API built on top of PyTorch.