However, in section 7.2.1, a dataset consisting of samples with labels 0 and 2 is created using the cifar10 variable.
cifar2 = [(img, label_map[label]) for img, label in cifar10 if label in [0, 2]]
I am assuming that the cifar10 variable here indicates the normalized cifar10 dataset. Hence, would it clearer to replace cifar10 with transformed_cifar10?
cifar2 = [(img, label_map[label]) for img, label in transformed_cifar10 if label in [0, 2]]
This will ensure that someone who is implementing these steps understands that the normalized data is now being used to train the NN.
In Chapter 7, the CIFAR10 dataset is initially loaded as -
Then, section 7.1.4 discusses the importance of normalizing the data. The transformed CIFAR10 dataset is loaded as -
However, in section 7.2.1, a dataset consisting of samples with labels 0 and 2 is created using the
cifar10
variable.I am assuming that the cifar10 variable here indicates the normalized cifar10 dataset. Hence, would it clearer to replace
cifar10
withtransformed_cifar10
?This will ensure that someone who is implementing these steps understands that the normalized data is now being used to train the NN.