o p im(@sddlmZddlmZddlmZddlmZddlm Z ddl m Z ddl m Z er2ddlmZgZe d d d d d  d#d$ddZe d dd dd  d#d$ddZe d dd dd  d#d$ddZe d dd d d  d#d$d!d"ZdS)%) annotations) TYPE_CHECKING)_C_ops)check_variable_and_dtype) LayerHelper)in_dynamic_or_pir_mode) deprecated)Tensorz2.4.0zpaddle.geometric.segment_sumz5paddle.incubate.segment_sum will be removed in future)ZsinceZ update_tolevelreasonNdatar segment_idsname str | NonereturncCtr t||dSt|dddt|dddtdit}|j|jd}|j|jd}|jd||d ||d d did |S)a  Segment Sum Operator. Sum the elements of input `data` which with the same index in `segment_ids`. It computes a tensor such that .. math:: out_i = \sum_{j \in \{segment\_ids_j == i \} } data_{j} where sum is over j such that `segment_ids[j] == i`. Args: data (Tensor): A tensor, available data type float32, float64, int32, int64. segment_ids (Tensor): A 1-D tensor, which have the same size with the first dimension of input data. Available data type is int32, int64. name (str, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`. Returns: Tensor, the Segment Sum result. Examples: .. code-block:: python >>> import paddle >>> data = paddle.to_tensor([[1, 2, 3], [3, 2, 1], [4, 5, 6]], dtype='float32') >>> segment_ids = paddle.to_tensor([0, 0, 1], dtype='int32') >>> out = paddle.incubate.segment_sum(data, segment_ids) >>> print(out) Tensor(shape=[2, 3], dtype=float32, place=Place(cpu), stop_gradient=True, [[4., 4., 4.], [4., 5., 6.]]) ZSUMXZfloat32Zfloat64int32int64 segment_pool SegmentIdsrr segment_sumdtyperrZOutZ SummedIdspooltypetypeZinputsZoutputsattrsN)r rrrrrlocalsZ"create_variable_for_type_inferencerZ append_opr rrhelperout summed_idsr)b/home/app/PaddleOCR-VL/.venv_paddleocr/lib/python3.10/site-packages/paddle/incubate/tensor/math.pyrs$/rzpaddle.geometric.segment_meanz6paddle.incubate.segment_mean will be removed in futurecCr)an Segment Mean Operator. Ihis operator calculate the mean value of input `data` which with the same index in `segment_ids`. It computes a tensor such that .. math:: out_i = \mathop{mean}_{j \in \{segment\_ids_j == i \} } data_{j} where sum is over j such that 'segment_ids[j] == i' and $n_i$ is the number of all index 'segment_ids[j] == i'. Args: data (tensor): a tensor, available data type float32, float64, int32, int64. segment_ids (tensor): a 1-d tensor, which have the same size with the first dimension of input data. available data type is int32, int64. name (str, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`. Returns: Tensor, the Segment Mean result. Examples: .. code-block:: python >>> import paddle >>> data = paddle.to_tensor([[1, 2, 3], [3, 2, 1], [4, 5, 6]], dtype='float32') >>> segment_ids = paddle.to_tensor([0, 0, 1], dtype='int32') >>> out = paddle.incubate.segment_mean(data, segment_ids) >>> print(out) Tensor(shape=[2, 3], dtype=float32, place=Place(cpu), stop_gradient=True, [[2., 2., 2.], [4., 5., 6.]]) ZMEANrrrrr segment_meanrrrrr N)r+r#r%r)r)r*r+ds$1r+zpaddle.geometric.segment_minz5paddle.incubate.segment_min will be removed in futurecCr)a Segment min operator. Calculate the minimum elements of input `data` which with the same index in `segment_ids`. It computes a tensor such that .. math:: out_i = \min_{j \in \{segment\_ids_j == i \} } data_{j} where min is over j such that `segment_ids[j] == i`. Args: data (tensor): a tensor, available data type float32, float64, int32, int64. segment_ids (tensor): a 1-d tensor, which have the same size with the first dimension of input data. available data type is int32, int64. name (str, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`. Returns: Tensor, the minimum result. Examples: .. code-block:: python >>> import paddle >>> data = paddle.to_tensor([[1, 2, 3], [3, 2, 1], [4, 5, 6]], dtype='float32') >>> segment_ids = paddle.to_tensor([0, 0, 1], dtype='int32') >>> out = paddle.incubate.segment_min(data, segment_ids) >>> print(out) Tensor(shape=[2, 3], dtype=float32, place=Place(cpu), stop_gradient=True, [[1., 2., 1.], [4., 5., 6.]]) ZMINrrrrr segment_minrrrrr N)r,r#r%r)r)r*r,s$0r,zpaddle.geometric.segment_maxz5paddle.incubate.segment_max will be removed in futurecCstr t||d}|St|dddt|dddtdit}|j|jd}|j|jd}|jd||d ||d d did |S)a Segment max operator. Calculate the maximum elements of input `data` which with the same index in `segment_ids`. It computes a tensor such that .. math:: out_i = \max_{j \in \{segment\_ids_j == i \} } data_{j} where max is over j such that `segment_ids[j] == i`. Args: data (tensor): a tensor, available data type float32, float64, int32, int64. segment_ids (tensor): a 1-d tensor, which have the same size with the first dimension of input data. available data type is int32, int64. name (str, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`. Returns: Tensor, the maximum result. Examples: .. code-block:: python >>> import paddle >>> data = paddle.to_tensor([[1, 2, 3], [3, 2, 1], [4, 5, 6]], dtype='float32') >>> segment_ids = paddle.to_tensor([0, 0, 1], dtype='int32') >>> out = paddle.incubate.segment_max(data, segment_ids) >>> print(out) Tensor(shape=[2, 3], dtype=float32, place=Place(cpu), stop_gradient=True, [[3., 2., 3.], [4., 5., 6.]]) MAXrrrrr segment_maxrrrrr N)r.r#)r rrr'r&r(r)r)r*r.s&0r.)N)r r rr rrrr ) __future__rtypingrZpaddlerZpaddle.base.data_feederrZpaddle.base.layer_helperrZpaddle.frameworkrZ paddle.utilsrr __all__rr+r,r.r)r)r)r*sT        ?A@