#4020·horovod

使用 Horovod 和 TensorFlow 发生崩溃

作者: mythZhu创建于 2024年2月5日更新于 2025年3月21日
标签bug

I did CPU training with Horovod + TensorFlow and launched it with OpenMPI. Horovod always crashed with the following errors when some workers didn't process any data and directly called hvd.join() to wait for other workers.

内容来源: horovod/horovod