nanograv / pint_pal

A long-lived repository for NANOGrav Pulsar Timing workflows and analysis.
MIT License
6 stars 16 forks source link

Hardcoded paths (noise + excised tim) #43

Open kmjc opened 11 months ago

kmjc commented 11 months ago

There are a few places where the directory structure and naming convention for the noise chains are hardcoded. There are a few separate expectations for what the path is and the files are called around the place, these are the ones I found.

in noise_utils.add_noise_to_model https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L277-L284

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L291

in noise_utils.analyze_noise https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L53

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L56

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L62

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L65

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L108-L111

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/noise_utils.py#L133

in utils.check_recentness_noise

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/utils.py#L1235-L1240

Also the path to the excised TOAs is hardcoded in utils.check_recentness_excision

https://github.com/nanograv/pint_pal/blob/50592396777518c49439a3e154c2d852efe883b7/src/pint_pal/utils.py#L1270-L1275

rossjjennings commented 1 month ago

Some of these, such as the use of chain_1.txt and pars.txt, are likely to be unproblematic, since the filenames are relatively standardized at the enterprise level. Others may cause problems, but it's likely we'll be changing the way we approach the noise analysis for NG20 anyway. We should revisit this once we have a better idea of how that will work.