chen-pin / ebs Goto Github PK

@chen-pin See

The Google Drive link to this spreadsheet is only accessible for JPL, not outside institutes.

In feature/mcmc branch the first four pytest functions in <test_ebs.py> fail due to missing config parameter.

When running "python example.py" it gives:

Traceback (most recent call last):
File "example.py", line 133, in
nemati2020_vvc6()
File "example.py", line 28, in nemati2020_vvc6
ref_json_filename=config['json_files']['ref_json'],
KeyError: 'ref_json'

Do not think we need to explicitly load those files since they can be passed directly to EXOSIMs. If a sweep over contrast or throughput is desired than those values can be updated and dumped to a file to be passed to EXOSIMs in which case the path in the exosims_param_dict would just need to be updated.

Make the format consistent by giving all input .csv files a header.

This should include a description of each file as well (with expected units)

Add documentation to the MCMC classes and potentially upload a sample notebook for how to run the MCMC mode.

Add a test to automatically cross check EBS outputs with expected outputs from the ESYWG ETC validation effort.