Hi!

I've got a question regarding pio train command.
I noticed that pio enriches class path with some libraries under
PIO_HOME/lib.

Under the hood it executes something like this:
spark-submit ..... --jar
file:engine.jar,file:pio/lib/spark/pio-data-elasticsearch1-assembly-0.11.0-incubating.jar,file:pio/lib/spark/pio-data-hbase-assembly-0.11.0-incubating.jar
...

We have a several engines on our pio platform.
For some reason I want to override that behaviour for one particular engine
and do not include pio-data libraries into training classpath, is there any
mechanism to do that?

Thanks,
George

Reply via email to