calibration
Create conda environment with Python 3.5 or later:
conda create -n {env_name} python=3.6
Activate conda environment:
conda activate {env_name}
conda install -c conda-forge openmpi
pip install -e .
-
Get the repository in your workspace, either on online or Maxwell cluster:
user@machine git clone https://git.xfel.eu/gitlab/dataAnalysis/calibration-services.git
Notebooks are available in the folder
calibration-services/notebooks
-
Via FastX client
Open a terminal and type the following commands
user@max-display: ssh exflgateway ssh exflonc<NN> #NN = 12 for eg. module load exfel calibration-services jupyter-notebook --port 8008 --no-browser
This will start a jupyter server on online cluster and a link will be provided that you can copy.
Now open another terminal and do the port forwarding
user@max-display: ssh -L 8008:localhost:8008 exflgateway -t ssh -L 8008:localhost:8008 exflonc<NN>
Open your regular browser and paste the link that you copied before the last step. Provided you have cloned the repository as shown in first step, the calibration-services folder will be visible in the jupyter notebook.
-
Via local computer
Open a terminal and type the following commands
user@localmachine$ ssh [email protected] ssh exflgateway ssh exflonc<NN> module load exfel calibration-services jupyter-notebook --port 8008 --no-browser --ip "*"
Now open another terminal and do the port forwarding
user@localmachine: ssh -L 8008:localhost:8008 [email protected] -t ssh -L 8008:exflonc<NN>:8008 exflgateway
Open your regular browser and paste the link that you copied before the last step.
-
From the hutch control room computers
Open a terminal and type the following commands
ssh exflonc<NN> module load exfel calibration-services jupyter-notebook --port 8008 --no-browser
Open another terminal and do the port forwarding
ssh -L 8008:localhost:8008 exflonc<NN>
Open your regular browser and paste the link that you copied before the last step.
On Maxwell cluster:
module load exfel calibration-services
Preferred way is to use jupyterhub instance running on https://max-jhub.desy.de/hub/login
Doing module load exfel calibration-services
once, a jupyter kernel calibration-services-kernel
will be created that can be used in jupyter notebooks.