buddi.dataset

Functions

get_dataset([input_tuple_order, ...])

Create a TensorFlow dataset from numpy arrays for training the BuDDI model.

train_validation_split(dataset[, val_size, seed])

Uses tensorflow native shuffle and take/skip to split a dataset into train and validation sets. This function will shuffle the dataset prior to splitting to ensure randomn split. However the index before and after shuffling will not be tracked so post splitting it will be very difficult to associate the train and val samples back to the original dataset/meta data. For this reason this function is only intended to be used for dividing the train data into train and validation sets to facilitate train and validation loss computation where there is no need to associate the train and validation samples back to the original dataset/meta data.

Modules

buddi3_dataset
buddi4_dataset
data_split
dataset_generator