Making multiple predictions for image not saved on disk

thanks @sgugger, section below?

but how to produce an ImageList from inmemory images ? Can I load it from array of tensors or similar?

(needs to be customimagelist? Load the whole dataset in RAM or Custom ImageList for "virtual" image patches/crops )

https://docs.fast.ai/tutorial.inference.html#A-classification-problem

" You can also do inference on a larger set of data by adding a test set . This is done by passing an ItemList to load_learner .

learn = load_learner(mnist, test=ImageList.from_folder(mnist/‘test’))

preds,y = learn.get_preds(ds_type=DatasetType.Test)
"