# Copyright 2020 Huawei Technologies Co., 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.
# ============================================================================
"""math Operations."""
from itertools import zip_longest
from collections import deque
import numpy as np
from mindspore.ops.composite.multitype_ops import _constexpr_utils as const_utils
from mindspore.common import dtype as mstype
from mindspore._checkparam import Validator as validator
from mindspore.ops.primitive import constexpr
from mindspore.ops import functional as F
from .. import operations as P
# count_nonzero
@constexpr
def _check_validate_axis(axis, name):
if isinstance(axis, (tuple, list)):
for idx, item in enumerate(axis):
validator.check_value_type("axis[%d]" % idx, item, [int], name)
axis = validator.check_value_type('axis', axis, [int, tuple, list], name)
return axis
@constexpr
def _check_validate_keepdims(keep_dims, name):
keep_dims = validator.check_value_type('keep_dims', keep_dims, [bool], name)
return keep_dims
[docs]def count_nonzero(x, axis=(), keep_dims=False, dtype=mstype.int32):
r"""
Count number of nonzero elements across axis of input tensor
Args:
x (Tensor): Input data is used to count non-zero numbers.
:math:`(N,*)` where :math:`*` means, any number of additional dimensions.
axis (Union[int, tuple(int), list(int)]): The dimensions to reduce. Only constant value is allowed.
Default: (), reduce all dimensions.
keep_dims (bool): If true, keep these reduced dimensions and the length is 1.
If false, don't keep these dimensions. Default: False.
dtype (Union[Number, mindspore.bool\_]): The data type of the output tensor. Only constant value is allowed.
Default: mindspore.int32
Returns:
Tensor, number of nonzero element. The data type is `dtype`.
Supported Platforms:
``Ascend`` ``GPU`` ``CPU``
Examples:
>>> # case 1: each value specified.
>>> x = Tensor(np.array([[0, 1, 0], [1, 1, 0]]).astype(np.float32))
>>> nonzero_num = ops.count_nonzero(x=x, axis=[0, 1], keep_dims=True, dtype=mindspore.int32)
>>> print(nonzero_num)
[[3]]
>>> # case 2: all value is default.
>>> nonzero_num = ops.count_nonzero(x=x)
>>> print(nonzero_num)
3
>>> # case 3: axis value was specified 0.
>>> nonzero_num = ops.count_nonzero(x=x, axis=[0,])
>>> print(nonzero_num)
[1 2 0]
>>> # case 4: axis value was specified 1.
>>> nonzero_num = ops.count_nonzero(x=x, axis=[1,])
>>> print(nonzero_num)
[1 2]
>>> # case 5: keep_dims value was specified.
>>> nonzero_num = ops.count_nonzero(x=x, keep_dims=True)
>>> print(nonzero_num)
[[3]]
>>> # case 6: keep_dims and axis value was specified.
>>> nonzero_num = ops.count_nonzero(x=x, axis=[0,], keep_dims=True)
>>> print(nonzero_num)
[[1 2 0]]
"""
const_utils.check_type_valid(F.dtype(x), mstype.number_type, 'input x')
axis = _check_validate_axis(axis, "count_nonzero")
keep_dims = _check_validate_keepdims(keep_dims, "count_nonzero")
const_utils.check_type_valid(dtype, mstype.number_type + (mstype.bool_,), 'dtype')
not_equal = P.NotEqual()
cast = P.Cast()
reduce_sum = P.ReduceSum(keep_dims)
nonzero_bool = not_equal(x, 0)
# ReduceSum only support float16 or float32 tensor.
nonzero_val = cast(nonzero_bool, mstype.float32)
nonzero_num = cast(reduce_sum(nonzero_val, axis), dtype)
return nonzero_num
# tensor dot
@constexpr
def _int_to_tuple_conv(axes):
"""
Converts ints to tuples in input axes, expected by most validation checks.
"""
for x in [0, 1]:
if isinstance(axes[x], int):
axes[x] = (axes[x],)
return axes
@constexpr
def _check_axes(axes, prim_name=None):
"""
Check for validity and type of axes passed to function.
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
validator.check_value_type('axes', axes, [int, tuple, list], "tensor dot")
if not isinstance(axes, int):
axes = list(axes) # to avoid immutability issues
if len(axes) != 2:
raise ValueError(f"{msg_prefix} dimension of 'axes' should be 2, but got 'axes': {axes}.")
axes = _int_to_tuple_conv(axes) # convert before length checks
if len(axes[0]) != len(axes[1]):
raise ValueError(f"{msg_prefix} first and second dim of 'axes' have to be the same size/length, "
f"but got 'axes': {axes}.")
if len(axes[0]) != len(set(axes[0])) or len(axes[1]) != len(set(axes[1])):
raise ValueError(f"{msg_prefix} 'axes' cannot have duplicating values, but got {axes}.")
return axes
@constexpr
def _typecheck_input(x1_type, x2_type, prim_name=None):
"""
Check input tensor types to be valid and confirm they are the same type.
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
const_utils.check_type_valid(x1_type, [mstype.float32, mstype.float16], 'x1')
const_utils.check_type_valid(x2_type, [mstype.float32, mstype.float16], 'x2')
if x1_type != x2_type:
raise TypeError(f"{msg_prefix} inputs must be the same type, but got x1_type: {x1_type} "
f"and x2_type: {x2_type}.")
@constexpr
def _axes_int_check(x1_shape, x2_shape, axes, prim_name=None):
"""
Convert from single int axes to 2d tuple if required
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
if isinstance(axes, int):
if axes < 0:
raise ValueError(f"{msg_prefix} axes must be at least 0, but got {axes}.")
if axes == 0:
# outer product, no input validation required
return [], []
if axes > len(x1_shape) or axes > len(x2_shape):
raise ValueError(f"{msg_prefix} axes cannot be greater than the length of x1_shape and x2_shape, "
f"but got axes: {axes}, x1_shape: {x1_shape}, x2_shape: {x2_shape}.")
x1_ind = tuple(range(len(x1_shape))[-1 * axes:])
x2_ind = tuple(range(len(x2_shape))[:axes])
axes = tuple((x1_ind, x2_ind))
axes = _int_to_tuple_conv(axes)
return axes
@constexpr
def _validate_axes(x1_shape, x2_shape, axes, prim_name=None):
"""
Checks for axes having the correct length according to input, for any value in axis
being out of range with given shape and also checking for compatible axes values
with given inputs.
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
shapes = [x1_shape, x2_shape]
# axis length check
for ix_input, x_axes in enumerate(axes):
axes_len = len(x_axes)
shape_dim_len = len(shapes[ix_input])
if axes_len > shape_dim_len:
raise ValueError(f"{msg_prefix} length of x_axes should be less than or equal to {shape_dim_len}, "
f"but got 'len(x_axes)': {axes_len}.")
# axis values range check
for ix_input, x_axes in enumerate(axes):
comp_shape = shapes[ix_input]
max_val = len(comp_shape) - 1
min_val = -1 * len(comp_shape)
for _, x_value in enumerate(x_axes):
if not min_val <= x_value <= max_val:
raise ValueError(f"{msg_prefix} value in axes should be in range: [{min_val}, {max_val}], "
f"but got {x_value}.")
# check axis value with input shape - both ways for axis valid
invalid_a = False
invalid_b = False
for i in range(len(axes[0])): # sizes already validated
if x1_shape[axes[0][i]] != x2_shape[axes[1][i]]:
invalid_a = True
if x1_shape[axes[0][i]] != x2_shape[axes[1][len(axes[0])-1-i]]:
invalid_b = True
if invalid_a and invalid_b:
raise ValueError(f"{msg_prefix} 'i' should exist such that 'x1_shape[axes[0][i]]' is equal to "
f"'x2_shape[axes[1][i]]' or 'x2_shape[axes[1][len(axes[0])-1-i]]', but got "
f"x1_shape: {x1_shape}, x2_shape: {x2_shape}, axes: {axes}.")
@constexpr
def _calc_new_shape(shape, axes, position=0):
"""
Calculate transpose and reshape parameters for input transformations,
'position' refers to whether tensor is first or second in the op.
"""
contraction_axes = tuple(i if i >= 0 else i + len(shape) for i in axes[position])
prod_contraction = int(np.prod([shape[i] for i in contraction_axes]))
free_axes = tuple(i for i in range(len(shape)) if i not in contraction_axes)
free_dims = tuple(shape[i] for i in free_axes)
prod_free = int(np.prod(free_dims))
transpose_perm = contraction_axes + free_axes if position else free_axes + contraction_axes
new_shape = (prod_contraction, prod_free) if position else (prod_free, prod_contraction)
return new_shape, transpose_perm, free_dims
[docs]def tensor_dot(x1, x2, axes):
"""
Computation of Tensor contraction on arbitrary axes between tensors `a` and `b`.
Contraction allows for the summation of products of elements of `a` and `b` on specified axes.
The same number of axes must be specified for both x1 and x2, and values must be within range
of number of dims of both `a` and `b`.
Selected dims in both inputs must also match.
axes = 0 leads to outer product
axes = 1 leads to normal matrix multiplication when inputs both 2D.
axes = 1 is the same as axes = ((1,),(0,) where both `a` and `b` are 2D.
axes = 2 is the same as axes = ((1,2),(0,1)) where both `a` and `b` are 3D.
Inputs:
- **x1** (Tensor) - First tensor in tensor_dot with datatype float16 or float32
- **x2** (Tensor) - Second tensor in tensor_dot with datatype float16 or float32
- **axes** (Union[int, tuple(int), tuple(tuple(int)), list(list(int))]) - Single value or
tuple/list of length 2 with dimensions specified for `a` and `b` each. If single value `N` passed,
automatically picks up last N dims from `a` input shape and first N dims from `b` input shape in order
as axes for each respectively.
Outputs:
Tensor, the shape of the output tensor is :math:`(N + M)`. Where :math:`N` and :math:`M` are the free axes not
contracted in both inputs
Raises:
TypeError: If `x1` or `x2` is not a Tensor.
TypeError: If `axes` is not one of the following: int, tuple, list.
Supported Platforms:
``Ascend`` ``GPU`` ``CPU``
Examples:
>>> input_x1 = Tensor(np.ones(shape=[1, 2, 3]), mindspore.float32)
>>> input_x2 = Tensor(np.ones(shape=[3, 1, 2]), mindspore.float32)
>>> output = ops.tensor_dot(input_x1, input_x2, ((0,1),(1,2)))
>>> print(output)
[[2. 2. 2]
[2. 2. 2]
[2. 2. 2]]
"""
shape_op = P.Shape()
reshape_op = P.Reshape()
transpose_op = P.Transpose()
matmul_op = P.MatMul(False, False)
# input validity checks
x1_shape = shape_op(x1)
x2_shape = shape_op(x2)
x1_type = F.dtype(x1)
x2_type = F.dtype(x2)
axes = _check_axes(axes, 'tensor_dot')
_typecheck_input(x1_type, x2_type, 'tensor_dot')
# input compatibility check & axes format update
axes = _axes_int_check(x1_shape, x2_shape, axes, 'tensor_dot')
_validate_axes(x1_shape, x2_shape, axes, 'tensor_dot')
x1_reshape_fwd, x1_transpose_fwd, x1_ret = _calc_new_shape(x1_shape, axes, 0)
x2_reshape_fwd, x2_transpose_fwd, x2_ret = _calc_new_shape(x2_shape, axes, 1)
output_shape = x1_ret + x2_ret # combine free axes from both inputs
# run tensor_dot op
x1_transposed = transpose_op(x1, x1_transpose_fwd)
x2_transposed = transpose_op(x2, x2_transpose_fwd)
x1_reshaped = reshape_op(x1_transposed, x1_reshape_fwd)
x2_reshaped = reshape_op(x2_transposed, x2_reshape_fwd)
mul_result = matmul_op(x1_reshaped, x2_reshaped)
final_result = reshape_op(mul_result, output_shape)
return final_result
@constexpr
def _check_invalid_input(x1_shape, x2_shape, prim_name=None):
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
if len(x1_shape) < 2 or len(x2_shape) < 2:
raise ValueError(f"{msg_prefix} inputs x1, x2 should have 'dimension >= 2',"
f"but got 'len(x1_shape)': ({len(x1_shape)}) and 'len(x2_shape)': ({len(x2_shape)}).")
@constexpr
def _typecheck_input_dot(x1_type, x2_type, prim_name=None):
"""
Check input tensor types to be valid and confirm they are the same type for dot and batch dot ops.
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
const_utils.check_type_valid(x1_type, [mstype.float16, mstype.float32], 'x1')
const_utils.check_type_valid(x2_type, [mstype.float16, mstype.float32], 'x2')
if x1_type != x2_type:
raise TypeError(f"{msg_prefix} inputs must be the same type, but got "
f"x1_type: {x1_type} and x2_type: {x2_type}.")
@constexpr
def _get_transpose_shape(x2_shape):
x2_shape_range = tuple(range(len(x2_shape)))
x2_shape_transpose = x2_shape_range[-2:-1] + x2_shape_range[:-2] + x2_shape_range[-1:]
return x2_shape_transpose
[docs]def dot(x1, x2):
"""
Computation a dot product between samples in two tensors.
Inputs:
- **x1** (Tensor) - First tensor in Dot op with datatype float16 or float32
The rank must be greater than or equal to 2.
- **x2** (Tensor) - Second tensor in Dot op with datatype float16 or float32
The rank must be greater than or equal to 2.
Outputs:
Tensor, dot product of x1 and x2.
Raises:
TypeError: If type of x1 and x2 are not the same.
TypeError: If dtype of x1 or x2 is not float16 or float32.
ValueError: If rank of x1 or x2 less than 2.
Supported Platforms:
``Ascend`` ``GPU`` ``CPU``
Examples:
>>> input_x1 = Tensor(np.ones(shape=[2, 3]), mindspore.float32)
>>> input_x2 = Tensor(np.ones(shape=[1, 3, 2]), mindspore.float32)
>>> output = ops.dot(input_x1, input_x2)
>>> print(output)
[[[3. 3.]]
[[3. 3.]]]
>>> print(output.shape)
(2, 1, 2)
>>> input_x1 = Tensor(np.ones(shape=[1, 2, 3]), mindspore.float32)
>>> input_x2 = Tensor(np.ones(shape=[1, 3, 2]), mindspore.float32)
>>> output = ops.dot(input_x1, input_x2)
>>> print(output)
[[[[3. 3.]]
[[3. 3.]]]]
>>> print(output.shape)
(1, 2, 1, 2)
>>> input_x1 = Tensor(np.ones(shape=[1, 2, 3]), mindspore.float32)
>>> input_x2 = Tensor(np.ones(shape=[2, 3, 2]), mindspore.float32)
>>> output = ops.dot(input_x1, input_x2)
>>> print(output)
[[[[3. 3.]
[3. 3.]]
[[3. 3.]
[3. 3.]]]]
>>> print(output.shape)
(1, 2, 2, 2)
>>> input_x1 = Tensor(np.ones(shape=[3, 2, 3]), mindspore.float32)
>>> input_x2 = Tensor(np.ones(shape=[2, 1, 3, 2]), mindspore.float32)
>>> output = ops.dot(input_x1, input_x2)
>>> print(output)
[[[[[3. 3.]]
[[3. 3.]]]
[[[3. 3.]]
[[3. 3.]]]]
[[[[3. 3.]]
[[3. 3.]]]
[[[3. 3.]]
[[3. 3.]]]]
[[[[3. 3.]]
[[3. 3.]]]
[[[3. 3.]]
[[3. 3.]]]]]
>>> print(output.shape)
(3, 2, 2, 1, 2)
"""
shape_op = P.Shape()
reshape_op = P.Reshape()
transpose_op = P.Transpose()
matmul_op = P.MatMul(False, False)
x1_shape = shape_op(x1)
x2_shape = shape_op(x2)
x1_type = F.dtype(x1)
x2_type = F.dtype(x2)
_typecheck_input_dot(x1_type, x2_type, 'dot')
_check_invalid_input(x1_shape, x2_shape, 'dot')
if len(x1_shape) > 2 or len(x2_shape) > 2:
x2_shape_transpose = _get_transpose_shape(x2_shape)
x2_transpose = transpose_op(x2, x2_shape_transpose)
x1_reshape = reshape_op(x1, (-1, x1_shape[-1]))
x2_reshape = reshape_op(x2_transpose, (x2_shape[-2], -1))
mul_result = matmul_op(x1_reshape, x2_reshape)
return reshape_op(mul_result, x1_shape[:-1] + x2_shape[:-2] + x2_shape[-1:])
return matmul_op(x1, x2)
@constexpr
def _get_batch_size(x1_shape, x2_shape, prim_name=None):
"""
Get batch sizes from two inputs
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
if len(x1_shape) < 2 or len(x2_shape) < 2:
raise ValueError(f"{msg_prefix} inputs x1, x2 should have 'dimension >= 2', "
f"but got 'len(x1_shape)': ({len(x1_shape)}) and 'len(x2_shape)': ({len(x2_shape)}).")
return x1_shape[0], x2_shape[0]
@constexpr
def _typecheck_input_batch_dot(x1_type, x2_type, prim_name=None):
"""
Check input tensor types to be valid and confirm they are the same type for batch dot ops.
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
const_utils.check_type_valid(x1_type, [mstype.float32], 'x1')
const_utils.check_type_valid(x2_type, [mstype.float32], 'x2')
if x1_type != x2_type:
raise TypeError(f"{msg_prefix} inputs must be the same type, but got x1_type: {x1_type} and "
f"x2_type: {x2_type}.")
@constexpr
def _check_axes_for_batch_dot(x1_shape, x2_shape, axes, prim_name=None):
"""
Check whether axes are valid and cast axes from tuple to list
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
if axes is None:
if len(x2_shape) == 2:
axes = [len(x1_shape) - 1, len(x2_shape) - 1]
else:
axes = [len(x1_shape) - 1, len(x2_shape) - 2]
if isinstance(axes, (list, tuple)):
if 0 in axes:
raise ValueError(f"{msg_prefix} axes cannot contain 0, but got axes: {axes}.")
if len(axes) != 2:
raise ValueError(f"{msg_prefix} length of axes must be equal to 2, but got {len(axes)}.")
if isinstance(axes, tuple):
axes = list(axes)
validator.check_value_type('axes[0]', axes[0], [int], 'batch_dot')
validator.check_value_type('axes[1]', axes[1], [int], 'batch_dot')
# Reverse if axis < 0
if axes[0] < 0:
axes[0] += len(x1_shape)
if axes[1] < 0:
axes[1] += len(x2_shape)
validator.check_non_negative_int(axes[0], 'reversed axes[0]', 'batch_dot')
validator.check_non_negative_int(axes[1], 'reversed axes[1]', 'batch_dot')
if axes[0] > len(x1_shape) or axes[1] > len(x2_shape):
raise ValueError(f"{msg_prefix} axes[0] must be less than or equal to len(x1_shape), "
f"and axes[1] must be less than or equal to len(x2_shape)."
f"But got axes: {axes}, x1_shape: {x1_shape}, x2_shape: {x2_shape}.")
elif isinstance(axes, int):
if axes == 0:
raise ValueError(f"{msg_prefix} axes should not equal to 0, but got {axes}.")
if axes < 0:
axes = [axes + len(x1_shape), axes + len(x2_shape)]
validator.check_non_negative_int(axes[0], 'reversed axes', 'batch_dot')
elif axes > len(x1_shape) or axes > len(x2_shape):
raise ValueError(f"{msg_prefix} axes cannot be greater than the length of x1_shape and x2_shape, "
f"but got axes: {axes}, x1_shape: {x1_shape}, x2_shape: {x2_shape}.")
else:
axes = [axes, axes]
else:
raise ValueError(f"{msg_prefix} type of axes must be one of those: int, tuple(int), list(int), "
f"but got {type(axes).__name__}.")
return axes
@constexpr
def _calc_new_shape_batchdot(shape, axes, position=0):
"""
Calculate transpose and reshape parameters for input transformations,
'position' refers to whether tensor is first or second in the op.
"""
axis = axes[position]
contraction_axes = tuple([axis])
prod_contraction = int(np.prod([shape[i] for i in contraction_axes]))
free_axes = tuple(i for i in range(1, len(shape)) if i not in contraction_axes)
free_dims = tuple(shape[i] for i in free_axes)
prod_free = int(np.prod(free_dims))
transpose_perm = contraction_axes + free_axes if position else free_axes + contraction_axes
transpose_perm = tuple([0]) + transpose_perm
new_shape = (prod_contraction, prod_free) if position else (prod_free, prod_contraction)
new_shape = tuple([shape[0]]) + new_shape
return new_shape, transpose_perm, free_dims
@constexpr
def _check_batch_size(x1_batch_size, x2_batch_size, prim_name=None):
"""
Check whether batch size of two inputs are the same
"""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
if x1_batch_size != x2_batch_size:
raise ValueError(f"{msg_prefix} both inputs x1, x2 should have the same batch sizes, but got "
f"x1_batch_size: {x1_batch_size} and x2_batch_size: {x2_batch_size}.")
@constexpr
def _get_output_shape(batch_size, x1_ret, x2_ret):
"""
Compute output shape for batch dot
"""
output_shape = tuple([batch_size]) + x1_ret + x2_ret
return output_shape
[docs]def batch_dot(x1, x2, axes=None):
"""
Computation of batch dot product between samples in two tensors containing batch dims.
.. math::
output = x1[batch, :] * x2[batch, :]
Inputs:
- **x1** (Tensor) - First tensor in Batch Dot op with datatype float32 and the rank of `x1` must be greater
than or equal to 2.
- **x2** (Tensor) - Second tensor in Batch Dot op with datatype float32. The datatype of `x2` should
be same as `x1` and the rank of `x2` must be greater than or equal to 2.
- **axes** (Union[int, tuple(int), list(int)]) - Single value or tuple/list of length 2 with dimensions
specified for `a` and `b` each. If single value `N` passed, automatically picks up last N dims from
`a` input shape and last N dimensions from `b` input shape in order as axes for each respectively.
Default: None.
Outputs:
Tensor, batch dot product of `x1` and `x2`.For example: The Shape of output
for input `x1` shapes (batch, d1, axes, d2) and `x2` shapes (batch, d3, axes, d4) is (batch, d1, d2, d3, d4),
where d1 and d2 means any number.
Raises:
TypeError: If type of x1 and x2 are not the same.
TypeError: If dtype of x1 or x2 is not float32.
ValueError: If rank of x1 or x2 less than 2.
ValueError: If batch dim used in axes.
ValueError: If len(axes) less than 2.
ValueError: If axes is not one of those: None, int, (int, int).
ValueError: If axes reversed from negative int is too low for dimensions of input arrays.
ValueError: If axes value is too high for dimensions of input arrays.
ValueError: If batch size of x1 and x2 are not the same.
Supported Platforms:
``Ascend`` ``GPU`` ``CPU``
Examples:
>>> x1 = Tensor(np.ones(shape=[2, 2, 3]), mindspore.float32)
>>> x2 = Tensor(np.ones(shape=[2, 3, 2]), mindspore.float32)
>>> axes = (-1, -2)
>>> output = ops.batch_dot(x1, x2, axes)
>>> print(output)
[[[3. 3.]
[3. 3.]]
[[3. 3.]
[3. 3.]]]
>>> x1 = Tensor(np.ones(shape=[2, 2]), mindspore.float32)
>>> x2 = Tensor(np.ones(shape=[2, 3, 2]), mindspore.float32)
>>> axes = (1, 2)
>>> output = ops.batch_dot(x1, x2, axes)
>>> print(output)
[[2. 2. 2.]
[2. 2. 2.]]
>>> print(output.shape)
(2, 3)
>>> x1 = Tensor(np.ones(shape=[6, 2, 3, 4]), mindspore.float32)
>>> x2 = Tensor(np.ones(shape=[6, 5, 4, 8]), mindspore.float32)
>>> output = ops.batch_dot(x1, x2)
>>> print(output.shape)
(6, 2, 3, 5, 8)
>>> x1 = Tensor(np.ones(shape=[2, 2, 4]), mindspore.float32)
>>> x2 = Tensor(np.ones(shape=[2, 5, 4, 5]), mindspore.float32)
>>> output = ops.batch_dot(x1, x2)
>>> print(output.shape)
(2, 2, 5, 5)
"""
transpose_op = P.Transpose()
batch_matmul_op = P.BatchMatMul()
squeeze_one_op = P.Squeeze(1)
squeeze_minus_one_op = P.Squeeze(-1)
# input validity checks
x1_shape = F.shape(x1)
x2_shape = F.shape(x2)
x1_dim_num = len(x1_shape)
x2_dim_num = len(x2_shape)
x1_type = F.dtype(x1)
x2_type = F.dtype(x2)
x1_batch_size, x2_batch_size = _get_batch_size(x1_shape, x2_shape, 'batch_dot')
_typecheck_input_batch_dot(x1_type, x2_type, 'batch_dot')
_check_batch_size(x1_batch_size, x2_batch_size, 'batch_dot')
axes = _check_axes_for_batch_dot(x1_shape, x2_shape, axes, 'batch_dot')
if x1_dim_num == 2:
x1 = F.expand_dims(x1, 1)
axes[0] += 1
if x2_dim_num == 2:
x2 = F.expand_dims(x2, 2)
x1_shape = F.shape(x1)
x2_shape = F.shape(x2)
x1_reshape_fwd, x1_transpose_fwd, x1_ret = _calc_new_shape_batchdot(x1_shape, axes, 0)
x2_reshape_fwd, x2_transpose_fwd, x2_ret = _calc_new_shape_batchdot(x2_shape, axes, 1)
output_shape = _get_output_shape(x1_batch_size, x1_ret, x2_ret)
x1_transposed = transpose_op(x1, x1_transpose_fwd)
x2_transposed = transpose_op(x2, x2_transpose_fwd)
x1_reshaped = F.reshape(x1_transposed, x1_reshape_fwd)
x2_reshaped = F.reshape(x2_transposed, x2_reshape_fwd)
# Batch matmal op part
mul_result = batch_matmul_op(x1_reshaped, x2_reshaped)
final_result = F.reshape(mul_result, output_shape)
# if the original dims are expanded, restore them from 3 to 2
if x1_dim_num == 2:
final_result = squeeze_one_op(final_result)
elif x2_dim_num == 2:
final_result = squeeze_minus_one_op(final_result)
return final_result
@constexpr
def _check_same_type(dtype1, dtype2):
return dtype1 == dtype2
@constexpr
def _max(*args):
"""Returns the maximum value."""
return max(*args)
@constexpr
def _min(*args):
"""Returns the minimum value."""
return min(*args)
@constexpr
def _infer_shape_rem(shape1, shape2, ndim1, ndim2, transpose_b):
"""Infers the shape of the last two dimensions after performing matmul."""
shape_rem = []
if ndim1 >= 2:
shape_rem.append(shape1[-2])
if transpose_b:
if ndim2 >= 2:
shape_rem.append(shape2[-2])
else:
if ndim1 >= 1:
shape_rem.append(shape2[-1])
return tuple(shape_rem)
@constexpr
def _check_matmul_shapes(shape1, shape2, prim_name=None):
"""Checks shape1 and shape2 are valid to perform matmul, and returns output shape after broadcasting."""
msg_prefix = f"For '{prim_name}', the" if prim_name else "The"
ndim1, ndim2 = len(shape1), len(shape2)
if ndim1 < 1 or ndim2 < 1:
raise ValueError(f"{msg_prefix} dimension of input operands must be at least 1, but got "
f"the length of shape1: {ndim1}, the length of shape2: {ndim2}.")
if ndim2 >= 2 and shape1[-1] != shape2[-2]:
raise ValueError(f"{msg_prefix} shape1[-1] should be equal to shape2[-2] when the length of shape2 "
f"is greater than or equal to 2, but got shape1[-1]: {shape1[-1]}, "
f"shape2[-2]: {shape2[-2]}.")
shape_out = deque()
for items in zip_longest(reversed(shape1[:-2]), reversed(shape2[:-2]), fillvalue=1):
max_size = max(items)
if any(item not in (1, max_size) for item in items):
raise ValueError(f"{msg_prefix} operands could not be broadcast together with shape1 {shape1} and "
f"shape2 {shape2}.")
shape_out.appendleft(max_size)
return tuple(shape_out)
@constexpr
def _tile_size(shape, out_shape, ndim):
"""Returns tile_size such that shape*tile_size = out_shape"""
size = [1] * ndim
for idx, (i, j) in enumerate(zip(shape, out_shape)):
if i != j:
size[idx] = j
return tuple(size)
@constexpr
def _check_need_broadcast(shape1, shape2):
"""Returns True if broadcast is necessary for batchmatmul."""
return shape1[:-2] != shape2[:-2]
def _expand(x, ndim):
"""Expand x to ndim from axis, which can be 0 or -1."""
while F.rank(x) < ndim:
x = F.expand_dims(x, 0)
return x
def _broadcast_to(x, shape_cur, shape_to, ndim_to):
"""Broadcasts x from shape_cur to shape_to."""
size = _tile_size(shape_cur, shape_to, ndim_to)
return F.tile(x, size)
[docs]def matmul(x1, x2, dtype=None):
"""
Returns the matrix product of two arrays.
Note:
Numpy arguments `out`, `casting`, `order`, `subok`, `signature`, and `extobj` are
not supported.
On GPU, the supported dtypes are np.float16 and np.float32.
On CPU, the supported dtypes are np.float16 and np.float32.
Args:
x1 (Tensor): Input tensor, scalar not allowed.
The last dimension of `x1` must be the same size as the second last dimension of `x2`.
And the shape of x1 and x2 could be broadcast.
x2 (Tensor): Input tensor, scalar not allowed.
The last dimension of `x1` must be the same size as the second last dimension of `x2`.
And the shape of x1 and x2 could be broadcast.
dtype (:class:`mindspore.dtype`, optional): defaults to None. Overrides the dtype of the
output Tensor.
Returns:
Tensor or scalar, the matrix product of the inputs. This is a scalar only
when both `x1`, `x2` are 1-d vectors.
Raises:
ValueError: If the last dimension of `x1` is not the same size as the
second-to-last dimension of `x2`, or if a scalar value is passed in.
ValueError: If the shape of `x1` and `x2` could not broadcast together。
Supported Platforms:
``Ascend`` ``GPU`` ``CPU``
Examples:
>>> # case 1 : Reasonable application of broadcast mechanism
>>> x1 = Tensor(np.arange(2*3*4).reshape(2, 3, 4), mindspore.float32)
>>> x2 = Tensor(np.arange(4*5).reshape(4, 5), mindspore.float32)
>>> output = ops.matmul(x1, x2)
>>> print(output)
[[[ 70. 76. 82. 88. 94.]
[ 190. 212. 234. 256. 278.]
[ 310. 348. 386. 424. 462.]]
[[ 430. 484. 538. 592. 646.]
[ 550. 620. 690. 760. 830.]
[ 670. 756. 842. 928. 1014.]]]
>>> print(output.shape)
(2, 3, 5)
>>> # case 2 : the rank of `x1` is 1
>>> x1 = Tensor(np.ones([1, 2]), mindspore.float32)
>>> x2 = Tensor(np.ones([2,]), mindspore.float32)
>>> output = ops.matmul(x1, x2)
>>> print(output)
[2.]
>>> print(output.shape)
(1,)
"""
# performs type promotion
dtype1 = F.dtype(x1)
dtype2 = F.dtype(x2)
if not _check_same_type(dtype1, dtype2):
x1 = x1.astype(mstype.float32)
x2 = x2.astype(mstype.float32)
ndim1_orig, ndim2_orig = F.rank(x1), F.rank(x2)
shape1_orig, shape2_orig = F.shape(x1), F.shape(x2)
transpose_b = ndim2_orig == 1
shape_backbone = _check_matmul_shapes(shape1_orig, shape2_orig, 'matmul')
# infers the shape of the output
shape_out = shape_backbone + _infer_shape_rem(shape1_orig, shape2_orig,
ndim1_orig, ndim2_orig, transpose_b)
x1 = _expand(x1, 2)
x2 = _expand(x2, 2)
if F.rank(x2) == 2:
if F.rank(x1) > 2:
x1 = F.reshape(x1, (-1, shape1_orig[-1]))
res = P.MatMul(False, transpose_b)(x1, x2)
else:
# broadcasts x1.shape[:-2] with x2.shape[:-2]
ndim_aligned = _max(ndim1_orig, ndim2_orig)
x1 = _expand(x1, ndim_aligned)
x2 = _expand(x2, ndim_aligned)
shape1_aligned, shape2_aligned = F.shape(x1), F.shape(x2)
x1 = _broadcast_to(x1, shape1_aligned[:-2], shape_backbone, ndim_aligned)
x2 = _broadcast_to(x2, shape2_aligned[:-2], shape_backbone, ndim_aligned)
res = P.BatchMatMul(False, transpose_b)(x1, x2)
if dtype is not None:
res = res.astype(dtype)
return F.reshape(res, shape_out)