jax.numpy.empty#

jax.numpy.empty(shape,dtype=None,*,device=None,out_sharding=None)[source]#

Create an empty array.

JAX implementation ofnumpy.empty(). Because XLA cannot create anun-initialized array,jax.numpy.empty() will always return an arrayfull of zeros.

Parameters:

• shape (Any) – int or sequence of ints specifying the shape of the created array.

• dtype (str |type[Any]|dtype |SupportsDType |None) – optional dtype for the created array; defaults to float32 or float64depending on the X64 configuration (seeDefault dtypes and the X64 flag).

• device (Device |Sharding |None) – (optional)Device orShardingto which the created array will be committed. This argument exists forcompatibility with thePython Array API standard.

• out_sharding (NamedSharding |PartitionSpec |None) – (optional)PartitionSpec orNamedShardingrepresenting the sharding of the created array (seeexplicit sharding for more details).This argument exists for consistency with other array creation routines across JAX.Specifying bothout_sharding anddevice will result in an error.

Returns:

Array of the specified shape and dtype, with the given device/sharding if specified.

Return type:

Array

See also

• jax.numpy.empty_like()

• jax.numpy.zeros()

• jax.numpy.ones()

• jax.numpy.full()

Examples

>>>jnp.empty(4)Array([0., 0., 0., 0.], dtype=float32)>>>jnp.empty((2,3),dtype=bool)Array([[False, False, False],       [False, False, False]], dtype=bool)