Tensor Creation Routines

Basic creation routines

mars.tensor.empty Return a new tensor of given shape and type, without initializing entries.
mars.tensor.empty_like Return a new tensor with the same shape and type as a given tensor.
mars.tensor.eye Return a 2-D tensor with ones on the diagonal and zeros elsewhere.
mars.tensor.identity Return the identity tensor.
mars.tensor.ones Return a new tensor of given shape and type, filled with ones.
mars.tensor.ones_like Return a tensor of ones with the same shape and type as a given tensor.
mars.tensor.zeros Return a new tensor of given shape and type, filled with zeros.
mars.tensor.zeros_like Return a tensor of zeros with the same shape and type as a given tensor.
mars.tensor.full Return a new tensor of given shape and type, filled with fill_value.

Creation from other data

mars.tensor.array Create a tensor.
mars.tensor.asarray Convert the input to an array.

Numerical ranges

mars.tensor.arange Return evenly spaced values within a given interval.
mars.tensor.linspace Return evenly spaced numbers over a specified interval.
mars.tensor.meshgrid Return coordinate matrices from coordinate vectors.
mars.tensor.mgrid Construct a multi-dimensional “meshgrid”.
mars.tensor.ogrid Construct a multi-dimensional “meshgrid”.

Building matrices

mars.tensor.diag Extract a diagonal or construct a diagonal tensor.
mars.tensor.diagflat Create a two-dimensional tensor with the flattened input as a diagonal.
mars.tensor.tril Lower triangle of a tensor.
mars.tensor.triu Upper triangle of a tensor.