The ESGF Compute end-user API provides an object-oriented climatology package to access large scale computational resources through the Web Processing Service interface standard.
Supported WPS version: 1.0.0
Conda is the preferred method of install.
conda install -c conda-forge -c cdat esgf-compute-api
or can be installed from source.
git clone https://github.com/ESGF/esgf-compute-api
cd esgf-compute-api
python setup.py install
You can start with the Getting Started which is a Jupyter notebook that can be ran in a JupyterLab instance.
or you can jump right into running some code. You can find instructions to retrieve a token here.
import cwt
# Create a variable from an OpenDAP url and the name of the variable.
tas = cwt.Variable('http://thredds/dap/test.nc', 'tas')
# Initialize the client with the url to the WPS endpoint and the Token/API key.
wps = cwt.WPSClient('http://localhost:8000/wps', api_key='<TOKEN>')
# Select the process to execut.
process = wps.process_by_name('CDAT.subset')
# Execut the process.
wps.execute(process, inputs=[tas])
# Wait for the process to complete, this will print status message to the console.
process.wait()
# Prints the output of the process which is either an instance of cwt.Variable, a list of cwt.Variable or a dict.
print(process.output)docker pull jasonb87/compute-api:2.2.0
docker run -it -p 8888:8888 jasonb87/compute-api:2.2.0
Open a browser to http://0.0.0.0:8888/lab and the password will be "esgf".
docker run -it jasonb87/compute-api:2.2.0 ipython
Jupyter Notebook cotnaining examples can be found here
Compatibility document can be found on here