Model Size bigger than expect

[bryanlinnan]

Hi Kaiyang, I trained mobilenetv2 which should be about 12M size, but the output model "model.pth.tar-50" is about 40M, is this normal? the output model seems to contain more than just model weights.

[KaiyangZhou]

that's a bit weird

so if you save the pure weights of mobilenetv2, it'd be around 9M

the 12M size might count the oprimizer's and scheduler's states

but anyway, which dataset are you using? I'd be curious if the large weights come from a heavy classification layer (say you have many many IDs)

[bryanlinnan]

my trainset data has 575 class, i think it's not the root cause, i wonder how can i save the pure '.pth' model?

[KaiyangZhou]

what do you mean by how to save the .pth model? isn't it as simple as using torch.save?

[bryanlinnan]

sorry, i thought the model size bigger than expectation is due to the model saved some other things, but it seems nothing else after i checked the code, so the question is still why the model size is such big? 500 classes will cause this?