You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Add support for SEOBNRv5HM and SEOBNRv5PHM as PMR piece (#10)
* Add support for SEOBNRv5HM and SEOBNRv5PHM for PMR piece
- added support for using SEOBNRv5HM and SEOBNRv5PHM as
the PMR piece in ESIGMA and ESIGMASur via pyseobnr
- pyseobnr kept as an optional dependency; code
uses lazy imports to avoid import errors if pyseobnr
waveform models aren't used and pyseobnr is not installed
- added exception handling for MR generation failure
- Update README for additional merger-ringdown choices
- Update ESIGMA usage notebook
---------
Co-authored-by: Prayush Kumar <prayush@users.noreply.github.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
***Inspiral piece (`InspiralESIGMAHM`):** It comes from a combination of post-Newtonian theory, self-force, and black hole perturbation theory.
32
32
***Plunge-merger-ringdown piece**: Assuming moderate starting eccentricities that decay by the late inspiral, we use the quasi-circular (QC) NR surrogate `NRSur7dq4` for the plunge-merger-ringdown piece for `ESIGMAHM`.
33
33
34
-
(**Note:** We also allow other QC `LALSuite`waveform models to be used as the plunge-merger-ringdown piece; see the optional argument `merger_ringdown_approximant` in the generation functions `get_imr_esigma_waveform` and `get_imr_esigma_modes`[here](https://github.com/gwnrtools/esigmapy/blob/master/esigmapy/generator.py). However, the default and the most tested choice is `NRSur7dq4`.)
34
+
(**Note:** We also allow other QC waveform models to be used as the plunge-merger-ringdown piece; see the optional argument `merger_ringdown_approximant` in the generation functions `get_imr_esigma_waveform` and `get_imr_esigma_modes`[here](https://github.com/gwnrtools/esigmapy/blob/master/esigmapy/generator.py). The models currently supported are: `NRSur7dq4`, `SEOBNRv4PHM` (via `LALSuite`), `SEOBNRv5HM`, `SEOBNRv5PHM` (via `pyseobnr`). However, the default and the most tested choice is `NRSur7dq4`.)
35
35
36
36
The full IMR waveform `ESIGMAHM` is produced by smoothly attaching the inspiral piece `InspiralESIGMAHM` to the plunge-merger-ringdown piece `NRSur7dq4`. This attachment is facilitated by `ESIGMAPy`.
* To avoid performing the above step in every new terminal session, either add the above command to your `.bashrc` file, or follow the instructions [here](http://gitlab.icts.res.in/akash.maurya/Installation-instructions/wikis/conda-tricks), replacing `PYTHONPATH` with `LAL_DATA_PATH`, to set this environment variable automatically on activating your `conda` environment.
74
74
75
75
### Installing `ESIGMAPy`
76
-
* Activate your `conda` environment and install `ESIGMAPy` by running: `pip install esigmapy`.
77
-
76
+
* Activate your `conda` environment and install `ESIGMAPy` by running: `pip install git+https://github.com/gwnrtools/esigmapy.git`.
77
+
* (Optional) If you want to use `SEOBNRv5HM` or `SEOBNRv5PHM` as the merger-ringdown piece in `ESIGMAHM`, install `ESIGMAPy` with `pyseobnr` option by running: `pip install "esigmapy[pyseobnr] @ git+https://github.com/gwnrtools/esigmapy.git"`
78
78
***
79
79
### Trying out `ESIGMAHM`
80
80
The instructions to generate `ESIGMAHM` waveforms and the various functionalities that it offers are detailed in [this tutorial notebook](https://github.com/gwnrtools/esigmapy/blob/master/notebooks/ESIGMA_tutorial.ipynb).
@@ -99,7 +99,7 @@ The instructions to generate `ESIGMAHM` waveforms and the various functionalitie
99
99
100
100
* Download the surrogate data files of `ESIGMASur`, which can be found [on the repo](https://github.com/gwnrtools/esigmapy/tree/master/esigmapy/surrogate/data). Next, set the shell environment variable `ESIGMASUR_DATA_PATH` to the directory where you keep these surrogate data files by running: `export ESIGMASUR_DATA_PATH="/path/to/ESIGMASur"`.
101
101
102
-
* Using the inspiral-only surrogate `InspiralESIGMASur` would not require any further dependencies. However, its hybridized IMR version `IMRESIGMASur` will require downloading the NR surrogate data file ([installation instructions above](https://github.com/gwnrtools/esigmapy/tree/master#installing-nrsur7dq4)).
102
+
* Using the inspiral-only surrogate `InspiralESIGMASur` would not require any further dependencies. However, its hybridized IMR version `IMRESIGMASur` will require downloading the NR surrogate data file ([installation instructions above](https://github.com/gwnrtools/esigmapy/tree/master#installing-nrsur7dq4)). Or, if you want to use `SEOBNRv5HM` or `SEOBNRv5PHM` as the merger-ringdown piece in `IMRESIGMASur`, add `pyseobnr` option during the installation: `pip install "esigmapy[surrogate,pyseobnr] @ git+https://github.com/gwnrtools/esigmapy.git"`.
103
103
104
104
### Trying out `ESIGMASur`
105
105
The usage instructions and the various functionalities of `ESIGMASur` are detailed in [this tutorial notebook](https://github.com/gwnrtools/esigmapy/blob/master/notebooks/ESIGMASur_tutorial.ipynb).
0 commit comments