问题描述

算力切分2切场景wide_deep模型训练失败：

环境信息

Hardware Environment(Ascend/GPU/CPU) / 硬件环境:
Ascend Training Solution 23.0.RC3.B080
CANN 7.0.RC1.B080
Ascend HDK 23.0.RC3.B080
Software Environment / 软件环境 (Mandatory / 必填):
-- MindSpore version (e.g., 2.0.0) :MindSpore 2.2.0 B070
-- Python version (e.g., Python 3.7.5) :python3.7.5
-- OS platform and distribution (e.g., Linux Ubuntu 16.04):openeuler22.03_arm

关联用例

vNPU_vir12_0004
/01xs5itoqrp/03210uq2ba2ln/03270v0lt2jej/03270v0lt2jfd/031l0v0m4asjqeb5/031l0v0m4b5c7eqh/031l0v0m4b5rqerh/031l0v0m4b5ujerm/031l0v0m4b5v4/031l0v0m4b60ierp/031l0v0m4bj48pkg/031l0v0m4bj4ppmc/

重现步骤

cd ../mindspore/official/recommend/Wide_and_Deep
python train_and_eval.py --data_path=/home/data/wide_deep_mindrecord/ --dataset_type=mindrecord --device_target=Ascend > train_log.txt 2>&1 &

预期结果

训练正常

日志/截图

[在这里上传图片]

备注

目前待看
董鸿儒 wx1150228