I'm using the ILSVRC2012 dataset and for that, I have the ground_truth_labels and the synset_words.txt. Now, I access this and download just the images containing animals.
I use a neural network to make a prediction and now, I want to calculate the classification error but I don't have the ground truth labels.
My question is: where can I find the ground truth labels for these images? The name of them appear like this
n00015388_12.JPEG,
n00015388_24.JPEG,
n00015388_119.JPEG
Is there any relationship between the name of the images and the files given by caffe? (synset_word.txt and val.txt)
What's the meaning of the numbers used to name the images?