Opened 2 months ago

Closed 2 months ago

#3258 closed help (fixed)

Analysing pp files on the RACC - python, iris, mo_pack

Reported by: gn907779 Owned by: um_support
Component: UM Tools Keywords: iris python mo_pack RACC
Cc: Platform: Other
UM Version: 10.6

Description

Hi,
I have been using a python script for converting pp files to netcdf. I have run this successfully on the met cluster for some time. I can't use the met cluster anymore. So I am trying to get this script to run on the RACC. When I use module load python2.7/canopy/2.1.8 the script goes the furthest and it complains that I require mo_pack to be installed. In order to run pip install Pawel told me to switch to "module load python2.7/anaconda/2018.12". pip install mo_pack gives "No matching distribution found for mo_pack" When I try the script again having loaded anaconda it doesn't get as far, saying "ImportError?: No module named iris.analysis". Hopefully there are people in Met processing UM files using python on the RACC. Where can I go for help?
Thanks,
Will

Change History (2)

comment:1 Changed 2 months ago by gn907779

I have found that iris is available from scitools.org.uk.
So I created my own python environment as described in the following knowledgebase article: https://research.reading.ac.uk/act/knowledgebase/python-on-the-academic-computing-cluster/
Then I executed the following commands:
conda install -c conda-forge iris
conda install -c conda-forge mo_pack
Running in my new modified python environment my script is now working.
This ticket can be closed.

comment:2 Changed 2 months ago by ros

  • Resolution set to fixed
  • Status changed from new to closed

Thanks for letting us know how you solved this.

Note: See TracTickets for help on using tickets.