Class TrainDataset

Dataset that will create train and validation batches from the given params.

the idea is to split the data into batches and allocate a 'valid_s' portion of them for validation. the position of the (continuous) validation block is w.r.t batches. E.g., for tot_size = 10, batch_s = 2, valid_idx=3, valid_s = 0.3 you get 4 + 1 train + valid batches: T T T V T

Parameters:

• tot_s - nr. of examples
• batch_s - batch size
• valid_s - fraction of data to allocate for validation
• valid_idx - batch index at which allocate validation data
• rng - numpy.RandomState used to shuffle batches or None (no shuffle) @return : (train_batches, valid_batches)

Overrides: object.__init__