diff --git a/docs/api/api_python/mindspore.ops.functional.rst b/docs/api/api_python/mindspore.ops.functional.rst index 5a4e8f34bcf..3bce8cace7e 100644 --- a/docs/api/api_python/mindspore.ops.functional.rst +++ b/docs/api/api_python/mindspore.ops.functional.rst @@ -647,4 +647,5 @@ Parameter操作函数 mindspore.ops.arange mindspore.ops.core mindspore.ops.count_nonzero + mindspore.ops.iou diff --git a/docs/api/api_python/ops/mindspore.ops.Dropout.rst b/docs/api/api_python/ops/mindspore.ops.Dropout.rst index df3e53eeba3..5d0d619b81a 100644 --- a/docs/api/api_python/ops/mindspore.ops.Dropout.rst +++ b/docs/api/api_python/ops/mindspore.ops.Dropout.rst @@ -5,22 +5,4 @@ Dropout是一种正则化手段，通过在训练中以 :math:`1 - keep\_prob` 的概率随机将神经元输出设置为0，起到减少神经元相关性的作用，避免过拟合。 - 参数： - - **keep_prob** (float) - 输入神经元保留概率，数值范围在0到1之间。例如，keep_prob=0.9，删除10%的神经元。默认值：0.5。 - - **Seed0** (int) - 算子层的随机种子，用于生成随机数。默认值：0。 - - **Seed1** (int) - 全局的随机种子，和算子层的随机种子共同决定最终生成的随机数。默认值：0。 - - 输入： - - **x** (Tensor) - Dropout的输入，任意维度的Tensor，其数据类型为float16或float32。 - - 输出： - - **output** (Tensor) - shape和数据类型与 `x` 相同。 - - **mask** (Tensor) - shape与 `x` 相同。 - - 异常： - - **TypeError** - `keep_prob` 不是float。 - - **TypeError** - `Seed0` 或 `Seed1` 不是int。 - - **TypeError** - `x` 的数据类型既不是float16也不是float32。 - - **TypeError** - `x` 不是Tensor。 - 更多细节请参考 :func:`mindspore.ops.dropout` 。 diff --git a/docs/api/api_python/ops/mindspore.ops.func_relu.rst b/docs/api/api_python/ops/mindspore.ops.func_relu.rst index 82c6b77aa0c..3fca3871239 100644 --- a/docs/api/api_python/ops/mindspore.ops.func_relu.rst +++ b/docs/api/api_python/ops/mindspore.ops.func_relu.rst @@ -1,7 +1,7 @@ mindspore.ops.relu ================== -.. py:function:: mindspore.ops.relu(input_x) +.. py:function:: mindspore.ops.relu(x) 线性修正单元激活函数（Rectified Linear Unit）。 @@ -14,12 +14,12 @@ mindspore.ops.relu 一般来说，与 `ReLUV2` 相比，此算子更常用。且 `ReLUV2` 会多输出一个掩码。 参数： - - **input_x** (Tensor) - relu的输入，shape： :math:`(N, *)` ，其中 :math:`*` 表示任意数量的附加维度， + - **x** (Tensor) - relu的输入，shape： :math:`(N, *)` ，其中 :math:`*` 表示任意数量的附加维度， 其数据类型为 `number `_。 返回： - Tensor，数据类型和shape与 `input_x` 相同。 + Tensor，数据类型和shape与 `x` 相同。 异常： - - **TypeError** - `input_x` 的数据类型不是数值型。 - - **TypeError** - `input_x` 不是Tensor。 + - **TypeError** - `x` 的数据类型不是数值型。 + - **TypeError** - `x` 不是Tensor。 diff --git a/docs/api/api_python/ops/mindspore.ops.func_relu6.rst b/docs/api/api_python/ops/mindspore.ops.func_relu6.rst index 0eb7ab7c04f..1a40fedaf20 100644 --- a/docs/api/api_python/ops/mindspore.ops.func_relu6.rst +++ b/docs/api/api_python/ops/mindspore.ops.func_relu6.rst @@ -1,7 +1,7 @@ mindspore.ops.relu6 ==================== -.. py:function:: mindspore.ops.relu6(input_x) +.. py:function:: mindspore.ops.relu6(x) 计算输入Tensor的ReLU（修正线性单元），其上限为6。 @@ -11,11 +11,11 @@ 返回 :math:`\min(\max(0,x), 6)` 元素的值。 参数： - - **input_x** (Tensor) - relu6的输入，shape： :math:`(N, *)` ，其中 :math:`*` 表示任意数量的附加维度，数据类型为float16或float32。 + - **x** (Tensor) - relu6的输入，shape： :math:`(N, *)` ，其中 :math:`*` 表示任意数量的附加维度，数据类型为float16或float32。 返回： - Tensor，数据类型和shape与 `input_x` 相同。 + Tensor，数据类型和shape与 `x` 相同。 异常： - - **TypeError** - 如果 `input_x` 的数据类型既不是float16也不是float32。 - - **TypeError** - 如果 `input_x` 不是Tensor。 + - **TypeError** - 如果 `x` 的数据类型既不是float16也不是float32。 + - **TypeError** - 如果 `x` 不是Tensor。 diff --git a/docs/api/api_python_en/mindspore.ops.functional.rst b/docs/api/api_python_en/mindspore.ops.functional.rst index 2ead6cd4bcc..a232482fd3d 100644 --- a/docs/api/api_python_en/mindspore.ops.functional.rst +++ b/docs/api/api_python_en/mindspore.ops.functional.rst @@ -95,6 +95,7 @@ Activation Functions :template: classtemplate.rst mindspore.ops.celu + mindspore.ops.dropout mindspore.ops.fast_gelu mindspore.ops.gumbel_softmax mindspore.ops.hardshrink @@ -643,8 +644,6 @@ Other Functions - Determine if two strings are equal. * - mindspore.ops.typeof - Get type of object. - * - mindspore.ops.iou - - Computes the intersection over union (IOU) or the intersection over foreground (IOF) for boxes.- .. msplatformautosummary:: :toctree: ops @@ -652,6 +651,6 @@ Other Functions :template: classtemplate.rst mindspore.ops.arange - mindspore.ops.batch_dot mindspore.ops.core mindspore.ops.count_nonzero + mindspore.ops.iou diff --git a/mindspore/python/mindspore/ops/function/math_func.py b/mindspore/python/mindspore/ops/function/math_func.py index c3b1a2de813..aa62b5a497e 100644 --- a/mindspore/python/mindspore/ops/function/math_func.py +++ b/mindspore/python/mindspore/ops/function/math_func.py @@ -5829,13 +5829,13 @@ def iou(anchor_boxes, gt_boxes, mode='iou'): and width are scaled by 0.2 internally. Args: - - **anchor_boxes** (Tensor) - Anchor boxes, tensor of shape (N, 4). "N" indicates the number of anchor boxes, - and the value "4" refers to "x0", "y0", "x1", and "y1". Data type must be float16 or float32. - - **gt_boxes** (Tensor) - Ground truth boxes, tensor of shape (M, 4). "M" indicates the number of ground - truth boxes, and the value "4" refers to "x0", "y0", "x1", and "y1". Data type must be float16 or float32. - - **mode** (string): The mode is used to specify the calculation method, - now supporting 'iou' (intersection over union) or 'iof' (intersection over foreground) mode. - Default: 'iou'. + anchor_boxes (Tensor): Anchor boxes, tensor of shape (N, 4). "N" indicates the number of anchor boxes, + and the value "4" refers to "x0", "y0", "x1", and "y1". Data type must be float16 or float32. + gt_boxes (Tensor): Ground truth boxes, tensor of shape (M, 4). "M" indicates the number of ground + truth boxes, and the value "4" refers to "x0", "y0", "x1", and "y1". Data type must be float16 or float32. + mode (string): The mode is used to specify the calculation method, + now supporting 'iou' (intersection over union) or 'iof' (intersection over foreground) mode. + Default: 'iou'. Returns: Tensor, the 'iou' values, tensor of shape (M, N), with the same data type as `anchor_boxes`. diff --git a/mindspore/python/mindspore/ops/function/nn_func.py b/mindspore/python/mindspore/ops/function/nn_func.py index 888234e6bc4..bc69f8f359a 100644 --- a/mindspore/python/mindspore/ops/function/nn_func.py +++ b/mindspore/python/mindspore/ops/function/nn_func.py @@ -1614,11 +1614,11 @@ def relu(x): `number `_. Returns: - Tensor of shape :math:`(N, *)`, with the same dtype and shape as the `input_x`. + Tensor of shape :math:`(N, *)`, with the same dtype and shape as the `x`. Raises: - TypeError: If dtype of `input_x` is not a number. - TypeError: If `input_x` is not a Tensor. + TypeError: If dtype of `x` is not a number. + TypeError: If `x` is not a Tensor. Supported Platforms: ``Ascend`` ``GPU`` ``CPU`` @@ -1645,15 +1645,15 @@ def relu6(x): It returns :math:`\min(\max(0,x), 6)` element-wise. Args: - x(Tensor) - Tensor of shape :math:`(N, *)`, where :math:`*` means, any number of + x (Tensor): Tensor of shape :math:`(N, *)`, where :math:`*` means, any number of additional dimensions, with float16 or float32 data type. Returns: - Tensor, with the same dtype and shape as the `input_x`. + Tensor, with the same dtype and shape as the `x`. Raises: - TypeError: If dtype of `input_x` is neither float16 nor float32. - TypeError: If `input_x` is not a Tensor. + TypeError: If dtype of `x` is neither float16 nor float32. + TypeError: If `x` is not a Tensor. Supported Platforms: ``Ascend`` ``GPU`` ``CPU`` @@ -2903,19 +2903,19 @@ def batch_norm(input_x, running_mean, running_var, weight, bias, training=False, Args: If `training` is False, `scale`, `bias`, `mean` and `variance` are Tensors. - input_x (Tensor) - Tensor of shape :math:`(N, C)`, with float16 or float32 data type. - running_mean (Tensor) - Tensor of shape :math:`(C,)`, has the same data type with `scale`. - running_var (Tensor) - Tensor of shape :math:`(C,)`, has the same data type with `scale`. - weight (Tensor) - Tensor of shape :math:`(C,)`, with float16 or float32 data type. - bias (Tensor) - Tensor of shape :math:`(C,)`, has the same data type with `scale`. + input_x (Tensor): Tensor of shape :math:`(N, C)`, with float16 or float32 data type. + running_mean (Tensor): Tensor of shape :math:`(C,)`, has the same data type with `scale`. + running_var (Tensor): Tensor of shape :math:`(C,)`, has the same data type with `scale`. + weight (Tensor): Tensor of shape :math:`(C,)`, with float16 or float32 data type. + bias (Tensor): Tensor of shape :math:`(C,)`, has the same data type with `scale`. If `training` is True, `scale`, `bias`, `mean` and `variance` are Parameters. - input_x (Tensor) - Tensor of shape :math:`(N, C)`, with float16 or float32 data type. - running_mean (Parameter) - Parameter of shape :math:`(C,)`, has the same data type with `scale`. - running_var (Parameter) - Parameter of shape :math:`(C,)`, has the same data type with `scale`. - weight (Parameter) - Parameter of shape :math:`(C,)`, with float16 or float32 data type. - bias (Parameter) - Parameter of shape :math:`(C,)`, has the same data type with `scale`. + input_x (Tensor): Tensor of shape :math:`(N, C)`, with float16 or float32 data type. + running_mean (Parameter): Parameter of shape :math:`(C,)`, has the same data type with `scale`. + running_var (Parameter): Parameter of shape :math:`(C,)`, has the same data type with `scale`. + weight (Parameter): Parameter of shape :math:`(C,)`, with float16 or float32 data type. + bias (Parameter): Parameter of shape :math:`(C,)`, has the same data type with `scale`. training (bool): If `training` is True, `mean` and `variance` are computed during training. If `training` is False, they're loaded from checkpoint during inference. Default: False. @@ -2925,7 +2925,7 @@ def batch_norm(input_x, running_mean, running_var, weight, bias, training=False, eps (float): A small value added for numerical stability. Default: 1e-5. Returns: - output_x (Tensor) - The same type and shape as the input_x. The shape is :math:`(N, C)`. + output_x (Tensor), The same type and shape as the input_x. The shape is :math:`(N, C)`. Raises: TypeError: If `training` is not a bool.