Jensen ZH Tan, Nicholas TF Tan and Maurice HT Ling*
School of Applied Sciences, Temasek Polytechnic, Singapore
*Corresponding Author: Maurice HT Ling, School of Applied Sciences, Temasek Polytechnic, Singapore.
Received: October 27, 2022; Published: November 15, 2022
Artificial neural network (ANN) is a computing system inspired by biological neural networks but recently, there is a move towards studying biological neural networks using neuronal simulations. Hence, ANN can be a tool to study biological neural networks. However, most ANN libraries only cater to one signal (equivalent to one neurotransmitter) and generally requires neurons to be organized into layers, which may not have direct biological equivalence. Here, we present Brainopy as a biologically relevant Python-based ANN library as it enables multiple neurotransmitters and allow each neuron to connect to any other neurons. The constructed neural network is persisted as an SQLite database file. Despite focusing on biological relevancy over computational efficiency, we built and simulated neural networks of up to 15000 neurons (within the neuronal complexity of Caenorhabditis elegans, a well-studied organism in neuroscience) using a retail laptop.
Keywords: Artificial Neural Network; Python Library; Biologically Relevant Representation; SQLite Persistence
Citation: Maurice HT Ling., et al. “Brainopy: A Biologically Relevant SQLite-Based Artificial Neural Network Library". Acta Scientific Computer Sciences 4.12 (2022): 13-22.
Copyright: © 2022 Maurice HT Ling., et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.