digicampipe

DigiCam pipeline based on ctapipe

The documentation can be found here: Digicampipe documentation

Installation

• install Anaconda
• git clone https://github.com/cta-sst-1m/digicampipe
• conda env create -f digicampipe/environment.yml
• source activate digicampipe
• pip install -e digicampipe
• pytest digicampipe/digicampipe

Long Form

Anaconda

You'll need Anaconda, so if you don't have it yet, just install it now. Follow the anaconda installation instructions. We propose to use the most recent version.

wget https://repo.continuum.io/archive/Anaconda3-5.0.0.1-Linux-x86_64.sh
bash ./Anaconda3-5.0.0.1-Linux-x86_64.sh

digicampipe

We propose to have a tidy place and clone digicampipe into a folder ctasoft/

mkdir ctasoft
cd ctasoft
git clone https://github.com/cta-sst-1m/digicampipe

To not mix up your anaconda root environment with digicampipe, we propose to make a so called environment, with all the dependencies in it.

Create new environment

conda env create -f digicampipe/environment.yml

Update your environment

This is usefull when new packages are added to the environment.yml

conda env update -f digicampipe/environment.yml

Activate the anaconda environment

source activate digicampipe

Please Note: When working with digicampipe, please always make sure you are really using the digicampipe environment. After source activate digicampipe your prompt should look similar to this this:

(digicampipe) username@host:~/ctasoft$

pip install -e digicampipe

Note for MacOS users at the moment the installation of the protozfitsreader package requires some manual work. Please type this command before running pip

export DYLD_LIBRARY_PATH=$DYLD_LIBRARY_PATH:<PATH_TO_YOUR_ANANCODA_LIBRARY_PATH>/lib/python3.6/site-packages

<PATH_TO_YOUR_ANANCODA_LIBRARY_PATH> is where your anaconda was installed.

Execute the tests

pytest digicampipe/digicampipe

All Tests

Some tests depend on astrometry.net (see below) and take long, so by default you might want to skip them. If you want to execute all tests, please do:

pytest -c all_tests.ini

Writing tests

To write tests please follow the instructions in the Pytest-doc. It is important that you add the test resources to the repo. To do this add the resource path to "package_data=" in the setup.py file.

Build the documentation with Sphinx (optional)

In the digicampipe directory run:

cd docs/
make html

This should create the documentation in digicampipe/docs/build/html. You can open the html files with your favorite web browser. To delete the documentation us:

make clean

Software usage

Mounting the data on your machine

You may want to mount the data from a remote onto your local machine. To do this you can use sshfs. Information about sshfs can be found here sshfs

Usage example :

sudo sshfs -o allow_other <username>@<remote_adress>:<remote_path> <local_path>

Example: Viewer

With digicampipe comes the example program digicam-view. To use it, just go call it like this:

dneise@lair:~/sst/data$ digicam-view SST1M01_20171030.066.fits.fz

And you might see something like this:

Pointing using lid CCD images

To be able to determine the pointing using stars reflections on the lid, astrometry.net is needed. If not installed locally, the web service will be used instead. To install astrometry.net:

sudo apt-get install astrometry.net

You will need to download catalogs too:

mkdir ~/astrometry-data
cd ~/astrometry-data
wget http://broiler.astrometry.net/~dstn/4200/wget.sh
chmod +x wget.sh
./wget.sh

Grab a beer as it will take a while ... Also, indexes 4203 and lower are probably not needed.

Then add the folowing line to /etc/astrmetry.cfg:

add_path ~/astrometry-data

That's it, you are ready to go.

template scan analysis

The purpose of the template scan analysis is to find the shape of the typical non-saturating pulse. To perform this analysis download the needed raw data files from somewhere and put them into someplace you like, then modify template_scan_analysis.sh so it finds the files. Afterwards call:

./template_scan_analysis.sh

... and wait. This analysis needs 3 cores, ~6GB memory and takes on my machine ~25minutes. Once the analysis is done, you can look at the results and perform the higher level anlysis with:

jupyter-notebook template_scan_analysis.ipynb