<![CDATA[Machine learning is a branch of artificial intelligence that focuses on learning itself (self-learning) or without having to be repeatedly programmed by humans. One of the new scientific fields in machine learning that has developed due to the development of GPU (Graphic Processing Unit) technology is deep learning. Currently, the use of the deep learning method can be one of the methods often used in research. One of them is the classification using the Convolutional Neural Network method. In this final project, the researcher has tested the most optimal classification performance with the fastest execution time by comparing the use of CPU (Central Processing Unit) with GPU or without GPU in considering the image in the Matlab software. The research was conducted in three stages, namely (1) the system design process (2) the execution program (3) parameter observation. According to the research results, testing using the GPU improves the accuracy and precision of the training and scanning process. However, GPU usage can take a longer training time when compared to a process without GPU. The scanning process without the GPU runs faster than when the GPU is on.]]>
Copyrights © 2021
