问题
I want to implement a custom mini-batch generator in Tensorflow 2.0 using tf.data.Dataset API. Concretely, I have image data, 100 classes with ~200 examples each. For each mini-batch, I want to randomly sample P classes, and K images from each class, for a total of P*K examples in a mini-batch (as described in the paper In Defense of the Triplet Loss for Person Re-Identification]).
I've been searching through documentation for tf.data.Dataset, but can't seem to find the right method. I've looked into the from_generator method, but it doesn't seem suitable for this, since it generates a whole dataset from scratch as I understood.
It seems to me that one way to do it would be to make a new class similar to BatchDataset which can be found in tf.data.Dataset source code, where I would somehow implement the logic, but I'm hoping for an easier solution to be honest.
来源：https://stackoverflow.com/questions/57570385/how-to-generate-custom-mini-batches-using-tensorflow-2-0-such-as-those-in-the-p