doppyo 0.0.2

Diagnostics and verification package

# doppyo

doppyo is a diagnostics/verification package for climate forecast systems. It is still in the very early stages of development.

Current functionality

Three modules are currently available

1. skill : functions for assessing one data set relative to another (usually model output relative to observation).

2. diagnostic : functions for computing various atmospheric and oceanic diagnostics.

3. utils : general support functions for the doppyo package.

Some test modules are also avaliable in doppyo.test. However, these are far from complete

Usage (If not using DCFP JupyterHub server)

1. Install anaconda if you have not already done so:

• Make your own directory in ‘/OSM/CBR/OA_DCFP/apps/’

mkdir /OSM/CBR/OA_DCFP/apps/<userid>

• A recent version of the anaconda installer is available in ‘/OSM/CBR/OA_DCFP/apps/anaconda_installer’. Run this and follow the prompts to install anaconda in the directory you have just created.

• The anaconda installer will add the anaconda path to your .bashrc (or equivalent). You will need to source this for the changes to be made:

source ~/.bashrc

• Check that anaconda python is now your default:

which python

2. Install doppyo:

• Clone the doppyo Bitbucket repository to your local machine. At your desired location, run:

git clone https://<userid>@bitbucket.csiro.au/scm/df/doppyo.git

• Replicate the doppyo environment locally. This will ensure that you have all libraries/packages required to run doppyo. Within the ‘requirements’ folder in your cloned repository, run:

conda env create -f doppyo_env_linux.yml

or

conda env create -f doppyo_env_mac.yml

depending on your os.

Alternatively, if updating an existing doppyo_env environment, run:

conda env update --file=doppyo_env_<os>.yml

• Activate the doppyo_env environment:

source activate doppyo_env

• Install the doppyo package. Within your cloned repository, run:

python setup.py install