Change Log

March 17th, 2026 Penny Wieser - V.1.0.19

Following reviewer comments, changed minimisation to use weighted least squares, rather than least squares. Note, for Neon, if you calibrated with the older versions of Diadfit, you should set minimise=’least_squares’ in step1, as we see a slightly larger shift in Ne peak position than for diads, and consistency with Neon lines during calibration is key. Realistically it only seems to affect the density by 0.001 g/cm3 anyway.

Jan 21st, 2026 Penny Wieser - V.1.0.16

Fixed bug where cosmic ray CRR code wasnt actually editing the file if it wasnt a headless_txt file - was just returning the original data Also fixed bug where if a spike is removed from the background, it was being carried through to the fit, resulting in a larger residual than it should have had. Points excluded from the background based on being N standard deviations away are now deleted before the fitting process - this stops lmfit from giving stupidly large errors. Thanks to Alex Bearden for spotting these issues!

Sept 4th, 2024 - Penny Wieser - V.1.0.3

Changed column names in H2O method from HW:LW_Trapezoid and HW:LW_Simpson to Water_to_HW_ratio_Trapezoid and Water_to_HW_ratio_Simpson Was also a bug where LW area was being populated in the HW area column in the output dataframe.

Publication version Penny Wieser - V.1.0.1

Updated code to allow for fitting of standards as a seperate groups replaced outdated simps with simpson for scipy. Also removed support for python 3.7, added 3.11, added scipy version.

Publication version Penny Wieser - V.1.0

Final tweaks to documentation and functions for publication

June 3rd, 2024 - Penny Wieser - V.0.91

Implemented parallel processing for Duan and Zhang (2006) EOS when doing monte carlo simulations Set fixed random state in numpy.

March 29th, 2024 - Penny Wieser - V.0.86

Was an issue in the math for the CO2-H2O EOS in the situation where H2O wasnt loss - mass ratio not calculated correctly. Caused convergence issues. Also fixed bug in MC code for FI - was using error from first sample for all samples (clear in output columns)

March 7th, 2024 - Penny Wieser - V.0.85

Duan and Zhang (2006) inverts for pressure from density using an adapted newton raphson method. At high densities, it was finding a negative root.

Built in a look up table to get starting pressure. Helps convergence. Now gives Nans for things in the look up table that yielded negative volumes.

Feb 4th, 2024 - Penny Wieser - V.0.82

Changes after review. Major restructuring of docs. When loading files, options for spectra_filetype, meta_file_ext, spectra_file_ext, old code with new versions will return error for these. No real way to make backwards compatible and get functionality for reviewer 2.

Also simplified Ne fitting to only use 1 function for baseline and peak, means can have 2 peaks for either pk1 or pk2

Dec 18th, 2023 - Penny Wieser - V.0.81

Added options to calculate mixed fluids pressure assuming H2O is lost, and not lost.

Nov 29th, 2023 - Penny Wieser - V.0.80

Added significant functionality for the CO2-H2O EOS of Duan and Zhang (2006)

Oct 25th, 2023 - Penny Wieser - v.0.78

Major tweak of the secondary peak finding function. Instead of using a sigma filter, uses prominence_filter, identifies peaks as if the highest peak is above the median of the two edges of the window. Old sigma_filter syntax will return error

June 23rd, 2023 - Penny Wieser

Changed function ‘plot_and_save_Ne_line_pickle’ to ‘generate_Ne_corr_model’

March 28th, 2023 - Penny Wieser

Changed ‘calculate_rho_gcm3_for_P_T’ to ‘calculate_rho_for_P_T’ for EOS calcs

March 22nd, 2023 - Penny Wieser

__version__ = ‘0.0.53’ Changed arguement in diad2_fit_config to be C13_prom not C13_abs_prom for consistency

Changed cent_generic to cent for secondary peak

Changelog started March 22nd, 2023, as syntax converging across different functions