Open Access

Table A.2

Selected architecture and hyperparameters.

Layer Type Layer Description
Input Size: 32x32x1
Convolutional No. of Filters: 16, Filter Size: 3, Activation: Relu, Padding: Same
Convolutional No. of Filters: 16, Filter Size: 3, Activation: Relu, Padding: Same
Max-Pooling Kernel Size: 2
Convolutional No. of Filters: 32, Filter Size: 3, Activation: Relu, Padding: Same
Convolutional No. of Filters: 32, Filter Size: 3, Activation: Relu, Padding: Same
Max-Pooling Kernel Size: 2
Fully Connected No. of Neurons: 1024, Activation: ReLu
Dropout Dropout Probability: 30%
Fully Connected No. of Neurons: 512, Activation: ReLu
Dropout Dropout Probability: 30%
Fully Connected No. of Neurons: 8, Activation: Softmax

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.