Hi the community

Do you mind introducing the task plugin of TonY, which is a framework to 
natively run
deep learning jobs on Apache Hadoop. It currently supports TensorFlow, PyTorch,
MXNet and Horovod.

TonY enables running either single node or distributed training as a Hadoop 
application.
This native connector, together with other TonY features, aims to run machine 
learning
 jobs reliably and flexibly.I think if can be supported into dolphinscheduler, 
it will make
people reuse the existing on-perm Hadoop cluster resources and dolphinscheduler 
will
also be a new workflow engine that covers the machine learning lifecycle jobs.

Feel free to discuss if having any ideas.

Best
Junfan Zhang

Reply via email to