This package contains a set of tools to load and visualize the contents of a set of openPMD files (typically, a timeseries).
The routines of openPMD-viewer can be used in two ways :
-
Use the Python API, in order to write a script that loads the data and produces a set of pre-defined plots.
-
Use the interactive GUI inside the Jupyter Notebook, in order to interactively visualize the data.
The notebooks in the folder tutorials/ demonstrate how to use both
the API and the interactive GUI. You can view these notebooks online
here.
You can also download and run these notebooks on your local computer
(when viewing the notebooks with the above link, click on Raw to be able to
save them to your local computer). In order to run the notebook on
your local computer, please install openPMD-viewer first (see
below), as well as wget (pip install wget).
If you wish to use the interactive GUI, the installation of
openPMD-viewer provides a convenient executable which automatically
creates a new pre-filled notebook and opens it in a
browser. To use this executable, simply type in a regular terminal:
openPMD_notebook
(This executable is installed by default, when installing openPMD-viewer.)
In order to install openPMD-viewer with conda, please install the Anaconda
distribution, and then type
conda install -c rlehe openpmd_viewer
If you cannot install openPMD-viewer with conda, the alternative
is to use pip. However, you need to first make sure that h5py is
installed on your local computer. This can be done for instance by
typing pip install h5py, though this may require you to install hdf5 separately.
Once h5py is installed, simply type
pip install openPMD-viewer
In addition, if you wish to use the interactive GUI, please type
pip install jupyter
If you wish to install the openPMD-viewer on a remote scientific
cluster, please make sure that the packages numpy, scipy and h5py
are available in your environment. This is typically done by a set of
module load commands (e.g. module load h5py) -- please refer to
the documentation of your scientific cluster.
Then type
pip install openPMD-viewer --user
Note: The package jupyter is only required for the interactive
GUI and thus it does not need to be installed if you are only using
the Python API. For NERSC users, access to Jupyter
notebooks is provided when logging to
https://ipython.nersc.gov.
We welcome contributions to the code! Please read this page for guidelines on how to contribute.