Currently, the env variable CLAIMED_DATA_PATH is mounted to data in the python working directory.
First, this does not work for R scripts because of the different working directory.
Second, users might want to mount different dirs that do not contain data, e.g. models or output.
Therefore, it would be easier to make it configureable similar to docker:
-v `pwd`/data:/home/docker/data -v `pwd`/model:/home/docker/model -v `pwd`/output:/home/docker/output
We might want to automate the different working dirs in Python and R.
Currently, the env variable
CLAIMED_DATA_PATHis mounted todatain the python working directory.First, this does not work for R scripts because of the different working directory.
Second, users might want to mount different dirs that do not contain data, e.g.
modelsoroutput.Therefore, it would be easier to make it configureable similar to
docker:We might want to automate the different working dirs in Python and R.