Train PANet on my own dataset

[karry-code]

Hi there,
I'm working on IDD dataset for performing semantic segmentation on it. I find PANet is a quite good model to carry out few shot learning for semantic segmentation and is suitable for my task. It would be really helpful if you could guide on how can I train PANet on my dataset (IDD dataset).

Thanks.

[kaixin96]

Hi @karry-code , the data loading code is quite VOC/COCO specific. If you want to use an external dataset, I suggest you checking out torchmeta. It has a cleaner abstraction for dataset class tailored for few-shot tasks. Thank you.

[karry-code]

Hi @kaixin96 , thanks a lot.

[roxhannes]

Hi @kaixin96 and @karry-code ,
I also would like to use PANet to segment some images of electrodes I am using for my research. I already have labeled images, but now I am struggeling how to train an own model. Can you help me how exactly it is possible to load external, non-published data, e.g. using torchmeta?
Thanks and bests,
Hannes

[Meoooww]

Did you successfully apply it?

[roxhannes]

No not yet, if you have any useful approach, I am happy to hear about it...