Generic Ray-based members¶
Objects which are storage format agnostic but require specific Ray implementation
are placed in modin.core.execution.ray.generic
.
Their purpose is to implement certain parallel I/O operations and to serve as a foundation for building storage format specific objects:
RayIO
– implements parallelto_csv()
andto_sql()
.GenericRayDataframePartitionManager
– implements parallelto_numpy()
.
- class modin.core.execution.ray.generic.io.io.RayIO¶
Base class for doing I/O operations over Ray.
- class modin.core.execution.ray.generic.partitioning.partition_manager.GenericRayDataframePartitionManager¶
The class implements the interface in PandasDataframePartitionManager.
- classmethod to_numpy(partitions, **kwargs)¶
Convert partitions into a NumPy array.
- Parameters
partitions (NumPy array) – A 2-D array of partitions to convert to local NumPy array.
**kwargs (dict) – Keyword arguments to pass to each partition
.to_numpy()
call.
- Return type
NumPy array