Exploring the effect of depth and width of CNN models on binary classification of dogs and cats

Convolutional neural networks (CNNs), one of the trendiest study areas in recent years, have garnered a lot of interest. Their performance is particularly notable in speech recognition, target identification, and picture recognition. The performance of a convolutional neural network model is particularly crucial because convolutional neural networks deal with problems that demand enormous amounts of data and have strict standards for result correctness. The convolutional neural network model's depth and width are the first two parameters that are examined in this paper, which is built on the Tensorflow framework. Next, the impact of depth on the accuracy of model's prediction is examined. The model's accuracy and loss are then used as a benchmark to assess its quality; at the same time, the performance of the model can be discussed in relation to the impact of width by selecting a different number of neurons with various convolutional kernel sizes; Finally, A combinational convolutional, the experiments' result show that, with a classification accuracy of up to 95.4%, the combinatorial convolutional neural network outperforms a model with single depth or width.