Overview over the files:

Current development:

• ets_fiber_assigner/netflow.py: fiber assignment tool based on a network flow algorithm

Mostly of historical interest:

• src/pyETS.cc: wrapper file describing the functionality exported to Python

• src/ets.*, src/ets_helpers.h: C++ implementation of the assigners and associated functionality

• Directory src/external/: C and C++ sources that were originally developed for the Planck simulation pipeline and can be re-used for ETS

Compiling/Installing:

Prerequisites

As far as possible, the package will install its dependencies automatically. However, it also depends on the "cobraOps" Python package, which currently has to be installed manually; see https://github.com/Subaru-PFS/ics_cobraOps/ for details.

The package allows to choose between the PULP package and the commercial (but free for academic use) Gurobi package for solving the network flow problem. One of those two needs to be installed and the appropriate flag needs to be set when calling the network solving routine observeWithNetflow().

Package installation

Simply do a python setup.py install or similar. pip install . should also work.

Demo code:

The file demo_netflow.py contains a very brief overview over the typical workflow, i.e. reading in target data from a file, converting target positions to x/y coordinates on the focal plane, determining which targets can be seen by which cobras, defining a cost function and planning several assignments

The input files containing the targets are ASCII and contain one target per line. The columns within a line contain the following information:

1. ID
2. R.A. [deg.]
3. Dec. [deg.]
4. Exposure Time [sec.]
5. Priority [1(highest) - 15(lowest)]
6. Magnitude [ABmag]. (optional)
7. Redshift (optional)
8. Object Type (optional)

An example file can be found in the data subdirectory.

Partial recalculation

It is possible to do a partial recalculation of an assignment after some visits have already been observed. This can become necessary for a large variety of reasons, a few examples being

• a fiber was discovered to be broken during the first visits
• some targets were not observed due to fiber collisions
• new, very high priority targets are added to the target lists
• it is discovered that some targets need more observation time than anticipated

In such a situation, the data structure must be set up in the same way as it would be for the full assignment calculation, but in addition a dictionary is passed to the assigner, which simply contains target IDs of already observed targets and the number of visits they have already been observed. This information is sufficient to compute the optimal assignment strategy for the remaining visits given the new parameters.

Example: An assignment is computed for a given target list and 5 visits. After three visits have been carried out, it is noticed that Cobra #0007 does not work. To compute the optimal strategy for the remaining two visits:

• make a dictionary of all science targets and the number of times they have been observed during the first three visits (excluding the one observed by Cobra #0007).
• remove Cobra #0007 from the list of active cobras
• run a fiber assignment task for 2 visits, passing the updated Cobra list and the above dictionary.

In case of any questions, please don't hesitate to contact me (martin@mpa-garching.mpg.de)!

Martin Reinecke