API Documentation

Merlin Loader for TensorFlow Constructors

Loader(dataset, batch_size[, shuffle, …])	Infinite generator used to asynchronously iterate through CSV or Parquet dataframes on GPU by leveraging an merlin.io.Dataset.
KerasSequenceValidater(dataloader)

Merlin Loader for TensorFlow Utility Functions

configure_tensorflow([memory_allocation, device])	Control the GPU memory allocation that is performed when using TensorFlow.
get_dataset_schema_from_feature_columns(…)	Maps from a list of TensorFlow feature_column to two lists.
HAS_GPU	bool(x) -> bool

Merlin Loader for PyTorch Constructors

Loader(dataset[, batch_size, shuffle, …])	This class creates batches of tensor.
DLDataLoader(dataset, batch_size, shuffle, …)	This class is an extension of the torch dataloader.

Merlin Loader Base Constructors

LoaderBase(dataset, batch_size, shuffle[, …])	Base class containing common functionality between the PyTorch and TensorFlow dataloaders.
ChunkQueue(dataloader, qsize[, num_parts, …])	This class takes partitions (parts) from an merlin.io.Dataset and concatenates them into a cudf dataframe “chunk.” This chunk is subsequently transformed into its tensor representation using the iterator’s transform.