ray.data.from_numpy_refs

ray.data.from_numpy_refs(ndarrays: ObjectRef[numpy.ndarray] | List[ObjectRef[numpy.ndarray]]) → MaterializedDataset

Creates a Dataset from a list of Ray object references to NumPy ndarrays.

The column name defaults to “data”.

Examples

>>> import numpy as np
>>> import ray
>>> arr_ref = ray.put(np.array([1]))
>>> ray.data.from_numpy_refs(arr_ref)  
shape: (1, 1)
╭───────╮
│ data  │
│ ---   │
│ int64 │
╞═══════╡
│ 1     │
╰───────╯
(Showing 1 of 1 rows)

Create a Ray Dataset from a list of NumPy array references.

>>> ray.data.from_numpy_refs([arr_ref, arr_ref])  
shape: (2, 1)
╭───────╮
│ data  │
│ ---   │
│ int64 │
╞═══════╡
│ 1     │
│ 1     │
╰───────╯
(Showing 2 of 2 rows)

Parameters:

ndarrays – A Ray object reference to a NumPy ndarray or a list of Ray object references to NumPy ndarrays.

Returns:

Dataset holding data from the given ndarrays.

DeveloperAPI: This API may change across minor Ray releases.