I have some questions if someone can answer me or guide me articles to understand them. I investigated different pre-trained model i.e. AlexNet, VGG, GoogLeNet, InceptionV3 and ResNet. I have retrained these models(In MATLAB2018b using single CPU) on my disease dataset and the retrained models are of following sizes:
AlexNet: 207.266MB
VGG: 407.981MB
GoogLeNet:22MB
Inception V3: 79.46MB
ResNet: 155MB
**Q1) Among all the size of the GoogLeNet and InceptionV3 is less among all? What can be the be the possible reasons? Usage of inception models or usage of 1x1 convolution filter?
Q2) Why the size of AlexNet is more than GoogLeNet and ResNet? Is it because of computation of gradient at each layer?**
The other question is related to understanding the training progress provided by MATLAB? Why the model struggle in the initial epoch? What are the possible options?
I will be really very thankful if anybody can help me in understanding any of these questions. Thanks in advance.
