Comparação da construção de redes neurais nas linguagens R e Python

Abstract

This report aims to make a comparison between Artificial Neural Networks (RNAs) written in R and Python languages on Windows 10 and Linux-Ubuntu 20.04 operating systems.The main metrics observed were: execution time in both languages at the different operating systems, considering the time for training and testing the network; the level of accuracy in the different operating systems will also be assessed.The tool used for the development was pycharm, together with the use of libraries keras, tensorflow and pandas. For the Python language, an environment was set up in miniconda 2 and for the R language the execution was by command line.The database used by the Neural Networks was taken from the Machine Learning Repository (UCI) and deals with the diagnosis of cancerous tumor.