FAQs

What if I don’t want to create a python virtual environment?

Sure you can avoid the use of virtual environment, if you want. In this case you have to follow all the step described Installation & updates except the environment sections.

Note that if you are installing ExoSim in your standard Python Environment, the system may ask you for administration privileges.

ExoSim is installed but not working: what can I do?

In our experience, mostly the errors raised after an failed run are of three different kinds:

  1. Python rasie an ImportError: No module named ####, for it cannot import the module used.

    Generally you can fix it installing the missing dependency, as

    pip install ####

    if the installation return errors, you may need to use administration privileges. For a Unix system you can use:

    sudo pip install ####

  2. Python raise an ImportError: No module named ExoSim.

    This means that ExoSim is not installed in the environment you are using. Try the installation procedure again or see the solution offered in What if I don’t want to create a python virtual environment?

What versioning system is used?

We are using a versioning system compliant to PEP440 standards. Given a version number as X.Y.Z

  • X is major modification identifier. Changing this number mean that we have refactored big part of the code. It’s not compatible with previous versions.

  • Y is minor modification. This number changes when we add functionalities or we change how part of the code work. A User will notice these variations and should check the documentation if some errors occur.

  • Z is patches identification. We fixed some problem or optimized something. Probably the user won’t notice the difference.

Other used versioning is in the form X.YbZ. The only difference is thar b stands for beta. You may also find c, meaning release candidate, or simply r, meaning release (this might be omitted) to indicate stable distributable versions.

How can I check which ExoSim version I’m using?

You can do this in multiple ways. The easiest way is to open the documentation (you already did) and look under the logo on the left panel or go to the changelog: there you will find all the versions listed. But this actually only refers to the directory you downloaded. To be sure that the version you are using is the same of the one you downloaded, run from the ExoSim Virtual Environment

pip show ExoSim

you will see all the installation details for ExoSim, including a Version line. Or you can check it from your python Virtual Environment

import ExoSim
ExoSim.__version__

or you can find it into the output files af an ExoSim module, looking for the ExoSim version metadata. Finally, you can even find that information inside the ExoSim.log file.

Be sure that version is the same reported in the documentation. If not, upgrade your installation with

pip install exosim --upgrade

if you installed ExoSim using pip (see install pip). If you used the source code from GitHub (see install git) go in your ExoSim directory, pull the last change and update your installation:

cd /your_path/ExoSim
git pull
pip install . --upgrade

and check again.

Tip

If you are using Anaconda Python, there must be a IDE listing all the installed package for your Virtual Environments and their versions.

How can I load HDF5 data into my code?

Once you have produced your dataset and it is stored into an .h5 file, you can use the data using the python package h5py. Assuming you data file is called data_file.h5, you can include it in your code as

import h5py

with h5py.File('data_file.h5', 'r+') as input_file:
    ...

Now the file can be navigated as a python dictionary. To read and use the data the user can refer to the documentation (https://docs.h5py.org/en/stable/high/dataset.html#reading-writing-data), but here is an example:

import h5py

with h5py.File('data_file.h5', 'r+') as input_file:
    data = input_file['first_level']['second_level']['dataset_name'][()]

This script navigates the file looking for the dataset called dataset_name that is under first_level/second_level and it loads all the dataset content into the data variable.

Load signals and tables

You can load the data stored into HDF5 file into their original python classes. In particular, you can cast a stored table into an QTable, using astropy.io.misc.hdf5.read_table_hdf5:

import h5py
from astropy.io.misc.hdf5 import read_table_hdf5

with h5py.File('data_file.h5', 'r+') as input_file:
    table_data = input_file['first_level']['table_group']
    table = read_table_hdf5(table_data)

where table_data is a dictionary loaded from the hdf5 that contains both the table and the table metadata, stored in the fle as .__table_column_meta__.

In the case of Signal class, you can use the exosim.output.hdf5.utils.load_signal:

import h5py
from exosim.output.hdf5.utils import load_signal

with h5py.File('data_file.h5', 'r+') as input_file:
    signal_group = input_file['first_level']['stored_signal_name']
    signal = load_signal(signal_group)