Part 1 lesson1.ipynb, 'arch' parameter question

data = ImageClassifierData.from_paths(PATH, tfms=tfms_from_model(arch, sz))
learn = ConvLearner.pretrained(arch, data, precompute=True)

Could we replace the ‘resnet34’ to other model? like ‘mobilenetv2’? I changed, but failed…

You can, has a bunch which you can call by default eg [‘resnet18’,

It’s also very easy to add mobilenetv2 too, you just need to find a version of it (like here install this, create an instance of this model, and then load the pretrained weights (model.load_state_dict(torch.load(‘PATHFORWEIGHTS’)))