o Me@sHddlZddlmmZddlmZddlmZejddfddZ dS)N)stream)ReduceOpTcCststj||||ddS|}|durdn|j}|tjkr(tj |d|d|S|tj kr7tj |d|d|S|tj krFtj |d|d|S|tjkrUtj|d|d|Std|)a Reduce a tensor over all ranks so that all get the result. As shown below, one process is started with a GPU and the data of this process is represented by its group rank. The reduce operator is sum. Through all_reduce operator, each GPU will have the sum of the data from all GPUs. .. image:: https://githubraw.cdn.bcebos.com/PaddlePaddle/docs/develop/docs/api/paddle/distributed/img/allreduce.png :width: 800 :alt: all_reduce :align: center Args: tensor (Tensor): The input Tensor. It also works as the output Tensor. Its data type should be float16, float32, float64, int32, int64, int8, uint8 or bool. op (ReduceOp.SUM|ReduceOp.MAX|ReduceOp.MIN|ReduceOp.PROD, optional): The operation used. Default value is ReduceOp.SUM. group (Group, optional): The group instance return by new_group or None for global default group. sync_op (bool, optional): Wether this op is a sync op. Default value is True. Returns: Return a task object. Examples: .. code-block:: python # required: distributed import paddle import paddle.distributed as dist dist.init_parallel_env() if dist.get_rank() == 0: data = paddle.to_tensor([[4, 5, 6], [4, 5, 6]]) else: data = paddle.to_tensor([[1, 2, 3], [1, 2, 3]]) dist.all_reduce(data) print(data) # [[5, 7, 9], [5, 7, 9]] (2 GPUs) F)opgroupsync_opuse_calc_streamNrrring_idzUnknown parameter: {}.) frameworkZ_in_legacy_dygraphr all_reduceidrSUMpaddleZ _legacy_C_opsZc_allreduce_sum_MAXZc_allreduce_max_ZMINZc_allreduce_min_ZPRODZc_allreduce_prod_ ValueErrorformat)Ztensorrrrrrr[D:\Projects\ConvertPro\env\Lib\site-packages\paddle/distributed/communication/all_reduce.pyr s<'        r ) r Zpaddle.fluid.frameworkZfluidr Z paddle.distributed.communicationrZ'paddle.distributed.communication.reducerr r rrrrs