#!/usr/bin/env python # -*- coding: utf-8 -*- # Copyright 1999-2020 Alibaba Group Holding Ltd. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from collections.abc import Iterable import numpy as np from ... import opcodes as OperandDef from ...serialize import ValueType, ListField from .core import TensorNoInput from .arange import arange from .empty import empty from .meshgrid import meshgrid class TensorIndices(TensorNoInput): _op_type_ = OperandDef.TENSOR_INDICES _dimensions = ListField('dimensions', ValueType.uint64) def __init__(self, dimensions=None, dtype=None, gpu=None, **kw): super().__init__(_dimensions=dimensions, _dtype=dtype, _gpu=gpu, **kw) @property def dimensions(self): return self._dimensions [docs]def indices(dimensions, dtype=int, chunk_size=None): """ Return a tensor representing the indices of a grid. Compute a tensor where the subtensors contain index values 0,1,... varying only along the corresponding axis. Parameters ---------- dimensions : sequence of ints The shape of the grid. dtype : dtype, optional Data type of the result. chunk_size : int or tuple of int or tuple of ints, optional Desired chunk size on each dimension Returns ------- grid : Tensor The tensor of grid indices, ``grid.shape = (len(dimensions),) + tuple(dimensions)``. See Also -------- mgrid, meshgrid Notes ----- The output shape is obtained by prepending the number of dimensions in front of the tuple of dimensions, i.e. if `dimensions` is a tuple ``(r0, ..., rN-1)`` of length ``N``, the output shape is ``(N,r0,...,rN-1)``. The subtensors ``grid[k]`` contains the N-D array of indices along the ``k-th`` axis. Explicitly:: grid[k,i0,i1,...,iN-1] = ik Examples -------- >>> import mars.tensor as mt >>> grid = mt.indices((2, 3)) >>> grid.shape (2, 2, 3) >>> grid[0].execute() # row indices array([[0, 0, 0], [1, 1, 1]]) >>> grid[1].execute() # column indices array([[0, 1, 2], [0, 1, 2]]) The indices can be used as an index into a tensor. >>> x = mt.arange(20).reshape(5, 4) >>> row, col = mt.indices((2, 3)) >>> # x[row, col] # TODO(jisheng): accomplish this if multiple fancy indexing is supported Note that it would be more straightforward in the above example to extract the required elements directly with ``x[:2, :3]``. """ from ..merge import stack dimensions = tuple(dimensions) dtype = np.dtype(dtype) raw_chunk_size = chunk_size if chunk_size is not None and isinstance(chunk_size, Iterable): chunk_size = tuple(chunk_size) else: chunk_size = (chunk_size,) * len(dimensions) xi = [] for ch, dim in zip(chunk_size, dimensions): xi.append(arange(dim, dtype=dtype, chunk_size=ch)) grid = None if np.prod(dimensions): grid = meshgrid(*xi, indexing='ij') if grid: grid = stack(grid) else: if raw_chunk_size is None: empty_chunk_size = None else: empty_chunk_size = (1,) + chunk_size grid = empty((len(dimensions),) + dimensions, dtype=dtype, chunk_size=empty_chunk_size) return grid