Using the pipeline remotely is the recommended method, in which case you don’t need to worry about software requirements.
However, for users who wish to go ahead with a local installation, we provide simple instructions in the current section.
The The Goodman Pipeline is completely written in Python 3.x and relies on several libraries like:
We do not recommend the installation of these libraries or the The Goodman Pipeline in your system since updates and upgrades may ruin it. We rather recommend the use of Virtual Environments. If you are not familiar with this term, please check the official documentation by visiting the links below:
Another option is to install Conda, a Virtual Environment Manager, or AstroConda, the same but for astronomers. Everything you need to know about installing both can be found in the link below:
Working with Virtual Environments¶
Virtual environments are a very useful tool, the main contribution of them being:
- Protection to the host environment
If you know nothing about them we recommend you to start in the Conda site.
For the purpose of this manual we will just say that a Virtual Environment lets you have a custom set of libraries/tools in one place, and most importantly is independent of your host system. Installation will not be discussed here but you can visit this link for information.
- Discover what environments exist in your system.
conda env list
Will print a list where the first column is the name.
- Activate (enter) the virtual Environment.
source activate <venv-name>
<venv-name>is the name of your virtual environment. Your shell’s prompt will change to:
(<venv-name>) [user@hostname folder-name]$
- Deactivate (leave) the virtual environment.
This time the prompt will change again to:
You may find that ccdproc and astroplan do not come with Astroconda. They are not available on any Conda channel either. That means that you will have to install them separately. You can do so by downloading the source files and installing them by hand, or simply activate your Virtual Environment and then install these two packages using pip with
pip install ccdproc astroplan
Setup for local installation¶
System installation is not recommended because it can mess things up specially in Linux and Mac OS. Before you proceed, make sure that your system has all the required libraries, as described in Requirements.
Once you have Python running and all the libraries installed either using Conda/AstroConda or not, you may download the last version available in the following address:
Before continuing, make sure that your Virtual Environment is active if this is the case. There are several ways of doing this but normally the command below should work:
$ source activate <my_environment_name>
<my_environment_name> is the name of your Virtual Environment (e.g.
Now you can finally install the The Goodman Pipeline. Download the file, decompress it, and enter the directory created during the file decompression. Test the installation by typing:
$ python setup.py test
If you have any errors, check the traceback. If you find difficulties carring on at this poing, you may contact us by opening a new issue or using the e-mail firstname.lastname@example.org.
If no error messages start popping up in your screen, you are good to carry on with the installation.
$ python setup.py install
This will install the pipeline in the currently active Python version.
If you have Virtual Environments, make sure that it is active. If not,
you can add the
--user option to install only for your user and avoid
needing root access.
Please cite: Pych, W., 2004, PASP, 116, 148
In terms of cosmic ray rejection we shifted to a non-python package because the results were much better compared to LACosmic’s implementation in Astropy. LACosmic was not designed to work with spectroscopy. Though since version 1.1.0 we shifted from Astropy to Astroscrappy’s implementation of LACosmic.
The latest version of the Goodman Spectroscopic Pipeline uses a modified version
dcr to help with the pipeline’s workflow. It is included under
goodman is the folder that will be created once you untar or unzip the latest
release of the The Goodman Pipeline.
The changes we made to DCR include deletion of all
which we don’t use in the pipeline. And addition of a couple of custom
keywords, such as:
GSP_FNAM, which stores the name of the file being
GSP_DCRR which stores the reference to the paper to cite.
You are still encouraged to visit the official Link. We remind again that users of the Goodman Pipeline should cite the DCR paper with the reference indicated above.
dcr is actually very simple.
Then simply type:
This will compile dcr and also it will create other files. The executable
binary here is
We have successfully compiled dcr right out the box in several platforms, such as:
- Ubuntu 16.04
- Centos 7.1, 7.4
- MacOS Sierra
- Solaris 11
Installing the DCR binary¶
This is a suggested method. If you are not so sure what you are doing, we
recommend you follow the steps shown below. If you are a more advanced user and
you want to do it your own way, all you have to achieve is to have the
executable binary in your
Open a terminal
In your home directory create a hidden directory
.bin(Home directory should be the default when you open a new terminal window)
Move the binary of your choice and rename it
dcr. If you compiled it, most likely it’s already called
dcrso you can ignore the renaming part of this step.
mv dcr.Ubuntu16.04 ~/.bin/dcr
mv dcr ~/.bin/dcr
$HOME/.bindirectory to your
$PATHvariable. Open the file
.bashrcand add the following line.
/home/myusernameis of course your home directory.
Close and reopen the terminal or load the