computer vision - Strange semantic segmentation results with UNet in Pytorch

Question

I was just trying to train UNet from scratch with a mammography dataset to detect tumor tissue in mammograms. after 35 training the model I got weird curves and bad results on the train and test set. I couldn't find any obvious mistake in my code, and everything seems correct.

Here is the link to implementation on GitHub: GitHub - hrnademi/Mammography

Further information about the dataset: Training images: 2400 images with the size of 256x256 png Test images: 600 images with the size of 256x256 png

Here are the curves:

Many thanks for your attention.

question from:https://stackoverflow.com/questions/65872009/strange-semantic-segmentation-results-with-unet-in-pytorch

Answer 1

Waitting for answers