Binary source combined with xallarap

[rpoleski]

Task to be done to have a model with binary source and their full orbital motion (BSPL_xallarap branch):

• [x] make sure .n_sources for xallarap models depends on whether q_source is defined or not
• [x] second source trajectory - shift by 180 in xi_u and change xi_semimajor_axis
• [x] ModelParameters._update_sources() - update and add calls to it
• [x] source mass ratio in ModelParameters
• [x] limit source mass ratio to > 0.
• [x] ModelParameters.source_2_parameters
• [x] make sure q_source cannot be mixed with, e.g., t_0_X
• [x] Currently the xallarap parameters of each source are not obtained directly (like, e.g., t_0_1) - do we want that? xi_a_2 is problematic. After discussion with @jenniferyee : No. We should give proper documentation only.
• [x] tests for FS case - code changed in commit 053a7603
• [x] Model.get_trajectory() (see also issue #94)
• [x] check all calls to Trajectory()
• [x] ModelParameters.xallarap_reference_position - define it in a standard way (including docstring)
• [x] ModelParameters.xallarap_reference_position - unit tests
• [x] print xallarap reference position while printing ModelParameters
• [x] properly print q_source
• [x] test Model.plot_trajectory()
• [x] test that printing parameters of both sources separately gives the same reference position
• [x] proper error message for mm.ModelParameters({'t_0': 1, 'u_0': 2, 't_E': 3, 'q_source': 1})
• [x] proper error message for xallarap parameters with rho_2 but without q_soruce
• [x] parameter_names.pdf
• [x] coverage of added code - modelparameters.py line 327
• [x] flake8, version number, sphinx, README.md

After merge:

• [ ] Example 16 - add new parameters to _set_default_parameters() and _check_fixed_parameters()

[rpoleski]

@jenniferyee What should be the parameter name for source mass-ratio? Possibilities I can think about:

1. q_source
2. q_xi
3. xi_q
4. something else?

[jenniferyee]

I think q_source.

Dr. Jennifer C. Yee (she/her)

Office: P-341 Center for Astrophysics | Harvard & Smithsonian 60 Garden St, MS-15 Cambridge, MA 02138

On Mon, Jul 24, 2023 at 11:38 AM Radek Poleski @.***> wrote:

@jenniferyee https://github.com/jenniferyee What should be the parameter name for source mass-ratio? Possibilities I can think about:

1. q_source
2. q_xi
3. xi_q
4. something else?

— Reply to this email directly, view it on GitHub https://github.com/rpoleski/MulensModel/issues/92#issuecomment-1648155384, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEK7UCO55UGFNXPFIMGMVB3XR2JFRANCNFSM6AAAAAA2VYQPRQ . You are receiving this because you were mentioned.Message ID: @.***>

[rpoleski]

Notes for myself on Trajectory:

• FitData.get_dataset_trajectory - called twice in that file for gradient calculation; that function should be written more consistently
• MagnificationCurve.trajectory - called only in that file; is MagnificationCurve only for a single source?
• Model.get_trajectory - called only in FitData.get_dataset_trajectory listed above