I just followed this great tutorial about how to quickly retrain ImageNet and make image classifiers using Tensorflow. I made the classifier, and it works well. From what I understand, Tensorflow partitions the provided dataset into training, test and validation by itself - or at least it does with this script. I've worked with sklearn in the past, and you can always find the accuracy of the model.
My question is, how can I find the accuracy percentage of the trained model in Tensorflow, specifically for image classifiers?
Thanks very much.