mars.tensor.zeros(shape, dtype=None, chunk_size=None, gpu=False, sparse=False, order='C')[source]

Return a new tensor of given shape and type, filled with zeros. Parameters ———- shape : int or sequence of ints

Shape of the new tensor, e.g., (2, 3) or 2.
dtype : data-type, optional
The desired data-type for the array, e.g., mt.int8. Default is mt.float64.
chunk_size : int or tuple of int or tuple of ints, optional
Desired chunk size on each dimension
gpu : bool, optional
Allocate the tensor on GPU if True, False as default
sparse: bool, optional
Create sparse tensor if True, False as default
order : {‘C’, ‘F’}, optional, default: ‘C’
Whether to store multi-dimensional data in row-major (C-style) or column-major (Fortran-style) order in memory.
out : Tensor
Tensor of zeros with the given shape, dtype, and order.

zeros_like : Return a tensor of zeros with shape and type of input. ones_like : Return a tensor of ones with shape and type of input. empty_like : Return a empty tensor with shape and type of input. ones : Return a new tensor setting values to one. empty : Return a new uninitialized tensor. Examples ——– >>> import mars.tensor as mt >>> mt.zeros(5).execute() array([ 0., 0., 0., 0., 0.]) >>> mt.zeros((5,), dtype=int).execute() array([0, 0, 0, 0, 0]) >>> mt.zeros((2, 1)).execute() array([[ 0.],

[ 0.]])
>>> s = (2,2)
>>> mt.zeros(s).execute()
array([[ 0.,  0.],
       [ 0.,  0.]])
>>> mt.zeros((2,), dtype=[('x', 'i4'), ('y', 'i4')]).execute() # custom dtype
array([(0, 0), (0, 0)],
      dtype=[('x', '<i4'), ('y', '<i4')])