Validation Accuracy doesn't increase.

Hello, I wonder if any of you who have used deep learning on matlab can help me to troubleshoot my problem. I don't understand why my model's validation accuracy doesn't increase. I used pre-trained AlexNet and My dataset just worked well in Python(PyTorch). I guess there is something problem with dataloader or image type(double, uint8, ...etc).

please help me. Thank you.

And here is my code.

[Sahar Salimpour]

Hello, i have the same problem. The validation accuracy is a fixed value and doesn't change at all(0.3), did you find any solution ?

Answer 1

[Jack Xiao]

reduce the initial learning rate to a smaller value such as 0.0001