Lab Assignment from AI for Beginners Curriculum.
Imagine say you wan develop one app for pet nursery wey go dey catalog all di pets. One beta feature for dis kain app na to fit sabi di breed from di photo. Dis one fit work well if we use neural networks.
You go need train one convolutional neural network to classify di different breeds of cats and dogs using Pet Faces dataset.
We go use di Oxford-IIIT Pet Dataset, wey get pictures of 37 different breeds of dogs and cats.
To download di dataset, use dis code snippet:
!wget https://thor.robots.ox.ac.uk/~vgg/data/pets/images.tar.gz
!tar xfz images.tar.gz
!rm images.tar.gzNote: Di Oxford-IIIT Pet Dataset pictures dey arranged by filename (e.g., Abyssinian_1.jpg, Bengal_2.jpg). Di notebook get code wey go help arrange di pictures into breed-specific subdirectories so classification go dey easier.
Start di lab by opening PetFaces.ipynb
You don solve one kind complex problem of image classification from scratch! Di classes plenty well well, but you still fit get reasonable accuracy! E still make sense to measure top-k accuracy, because e easy to confuse some classes wey no too clear even for human beings.
Disclaimer:
Dis dokyument don use AI translation service Co-op Translator take translate am. Even though we dey try make sure say e correct, abeg sabi say automatic translation fit get mistake or no dey accurate well. Di original dokyument for im native language na im you go take as di correct one. For important information, e good make you use professional human translation. We no go fit hold responsible for any misunderstanding or wrong interpretation wey fit happen because you use dis translation.