Özet:

In this study, various experiments have been performed via deep learning models based on CNN (Convolutional Neural Networks), ResNet (Residential Energy Services Network) and NIN (Network In Network) approaches and their performances on various datasets have been investigated. The deep learning models were named as E-Model, R-Model and A-Model, respectively. The deep learning models were trained with CIFAR-10 dataset on a machine having only CPU (Central Processing Unit) and a machine having both CPU and GPU (Graphical Processing Unit). On the machine having only CPU, the traning time of the R-Model, A-Model and E-Model were approximately 415 hours, 129 hours and 3.5 hours, respectively. The percentage correct values on the validation data set were %82.76, %87.64 ve %83.47, respectively. On the machine having both CPU and GPU, the traning time of the R-Model, A-Model and the E-Model were approximately 4.45 hours, 2.20 hours and 1.82 hours, respectively. The percentage correct values on the validation data set were %82.61, %87.95 ve %82.43, respectively. The experimental results for the other data sets were obtained by training the models on the machine having both CPU and GPU. The structures of the constructed deep learning models, the parameters used for the training, the obtained confusion matrices for the validation data, the accuracy and loss graphics are given in detail.

Anahtar Kelimeler: