mindspore.nn.PipelineCell

class mindspore.nn.PipelineCell(network, micro_size, stage_config=None, segment_config=None)[source]

Slice MiniBatch into finer-grained MicroBatch for use in pipeline-parallel training.

Note

micro_size must be greater or equal to pipeline stages.
The api will be deprecated, please use the api mindspore.parallel.nn.Pipeline instead.

Parameters

network (Cell) – The target network to wrap.
micro_size (int) – MicroBatch size.
stage_config (dict, optional) – The stage configuration for each cell's execution in pipeline parallel. Default None.
segment_config (dict, optional) – The segment configuration for each cell's execution in pipeline parallel. Default None.

Supported Platforms:: Ascend GPU

Examples

>>> import mindspore.nn as nn
>>> # Define the network structure of LeNet5. Refer to
>>> # https://gitee.com/mindspore/docs/blob/r2.7.0/docs/mindspore/code/lenet.py
>>> net = LeNet5()
>>> net = nn.PipelineCell(net, 4)