This repository contains the codebase and proof-of-work for my Final-Year Thesis. As mentioned in the repository description, the materials here demostrates the training, testing and implementation of Artificial Neural Networks on microcontrollers with TensorFlow/ TensorFlow Lite. These were part of the work done to evaluate the technical and commercial feasibility of using Machine Learning on Battery Management Systems for temperature estimation.
Although the materials here may be used freely by any interested parties, explanations are minimal since it was not within the primary objectives of the thesis to educate readers on Artificial Neural Networks. Additionally, since the Python scripts found here are only meant for data visualisation and serve as a proof-of-work, the datasets used have been deliberately left out for confidentiality purposes.
Interested parties may find the following links/tutorials useful: