Open cqulilujia opened 7 months ago
Describe the bug/ 问题描述 (Mandatory / 必填) 跑模型推理时,nn.Dense层会卡住,若在Dense层之前打印tensor信息或断点调试,可正常继续运行;设置mindspore.set_context(pynative_synchronize=True)也可正常继续运行
mindspore.set_context(pynative_synchronize=True)
Hardware Environment(Ascend/GPU/CPU) / 硬件环境: CPU
Ascend
GPU
CPU
Software Environment / 软件环境 (Mandatory / 必填): -- MindSpore version 2.2.11 -- Python version :python 3.9.19 -- OS platform and distribution:Linux Ubuntu 18.04
Excute Mode / 执行模式 (Mandatory / 必填)(PyNative/Graph):
PyNative
Graph
Please delete the mode not involved / 请删除不涉及的模式: /mode pynative
To Reproduce / 重现步骤 (Mandatory / 必填) Steps to reproduce the behavior: 开发过程中发现,非必现
Screenshots/ 日志 / 截图 (Mandatory / 必填) If applicable, add screenshots to help explain your problem.
您好,目前MindSpore已升级至2.3.rc1版本,修复了部分并发问题,建议您升级后再次尝试
Describe the bug/ 问题描述 (Mandatory / 必填) 跑模型推理时,nn.Dense层会卡住,若在Dense层之前打印tensor信息或断点调试,可正常继续运行;设置
mindspore.set_context(pynative_synchronize=True)
也可正常继续运行Hardware Environment(
Ascend
/GPU
/CPU
) / 硬件环境: CPUSoftware Environment / 软件环境 (Mandatory / 必填): -- MindSpore version 2.2.11 -- Python version :python 3.9.19 -- OS platform and distribution:Linux Ubuntu 18.04
Excute Mode / 执行模式 (Mandatory / 必填)(
PyNative
/Graph
):To Reproduce / 重现步骤 (Mandatory / 必填) Steps to reproduce the behavior: 开发过程中发现,非必现
Screenshots/ 日志 / 截图 (Mandatory / 必填) If applicable, add screenshots to help explain your problem.