Increasing training accuracy of U-Net segmentation model

General Discussion
,
#1

I was working with segmentation using u-net and MobileNet.

While I training with input 256*256 it had an output with Val loss: 0.044 (In this time dense layer was 256, 128, 64, 32 with a learning rate of 0.0001).

But I needed to modify input for deployment to 128*128. But then the val_loss increased drastically around 0.052 which is not acceptable.

What should I take care of?

#3

Hi @MSI

Welcome to the TensorFlow Forum!

Please share minimal reproducible code to replicate and understand the issue before helping you to fix this issue. Thank you.