40+% faster training with a scikit-learn-like API for numpy arrays

[muellerzr (Zachary Mueller)]

@oguiza did you attempt to use this method? It’ll speed it up some in native fastai (specifically the fake_dl). You should find it’s very similar to your PyTorch numbers

Speeding Up fastai2 Inference - And A Few Things Learned

I finally found an answer that makes a bit more sense. When doing one image I can get within 1ms by doing: %%timeit test = dls.test_dl([fnames[0]]) with test.fake_l.no_multiproc(): out = next(iter(test)) Which this makes much more sense on how to go about doing it using the library (and also can be a little bit faster than doing it my way.)