Model accuracy drops after a few epoch when training with custom dataset !

[ds-jpg]

Hi @plemeri

Thank you for your work.

I tried training the swin b model from scratch with my custom dataset which is having 420000 images of humans , products , cars . etc (210000 * 2 Horizontal Flip ) , after a few epoch of training it starts giving weird output and the accuracy drops and the output from the model becomes very poor.

I had changed the batch size to 8 in order to train the model little quicker but the results started to get very horrible after a few epochs. Can you please tell me why this is occurring ?

Please see the attached input and output images for a better understanding of the problem.
image-compare.pdf

Thanks :)

[plemeri]

Hello @ds-jpg

First of all, by just seeing the results cannot tell the problem exactly.
But I think the issue came from the training and inference image size difference.

I recommend using another configuration which we used for real-world scenario with composite datasets.
Plus_Ultra.yaml

Thank you.