Outputs

Fitspy provides two types of outputs:

those related to the models,
and those related to the fits (parameters of the fitted models and statistics).

Outputs models

The output models consist of .json files in which all parameters of the spectrum and spectra objects have been saved so as to be able:

either to recreate the previously obtained results identically

or to apply the first spectrum model to other data sets (without considering the fname value):

{
   "0": {
      "fname": "../examples/data/spectra_2/spectrum_2_1.txt",
      "range_min": 200.0,
      "range_max": 957.299,
      "normalize": true,
      "normalize_range_min": 670.0,
      "normalize_range_max": 700.0,
      "models_labels": ["", "Peak IV", "Peak II", "", "", "Peak I", "Peak III", "", ""],
      "bkg_model": null,
      "fit_method": "Leastsq",
      "fit_negative": true,
      "max_ite": 200,
      "attached": true,
      "baseline": {
         "points": [[201.1400035483871, 954.8547838709676], [164.91378720565154, 228.197032967033]],
         "mode": "Linear",
         "order_max": 2,
         "distance": 100,
         "sigma": 0,
         "attached": true,
         "is_subtracted": true
      },
      "models": {
         "0": {
            "Lorentzian": {
               "ampli": {
                  "value": 377.8408958125738,
                  "min": 0,
                  "max": Infinity,
                  "vary": 1,
                  "expr": null
               },
               "x0": {
                  "value": 342.94548419455225,
                  "min": 322.195,
                  "max": 362.195,
                  "vary": 1,
                  "expr": null
               },
               "fwhm": {
                  "value": 2.643513972868272,
                  "min": 0,
                  "max": 200,
                  "vary": 1,
                  "expr": null
               }
            }
         },
         "1": {
            "Lorentzian": {
               "ampli": {
                  "value": 98.30555093353449,
                  "min": 0,
                  "max": Infinity,
                  "vary": 1,
                  "expr": null
               },
               ...
         },
         ...

Fit results

Once fitted, a spectra data set can generate several files related to the fitting statistics and the fitted parameters.

In a ‘results’ directory (to be specified by the user), the statistics files are built when clicking on Save Results in the GUI, according to the following model: {spectrum basename}**_stats.txt** file).

The same for the fitted parameters attached to each spectrum: {spectrum basename}**.csv** file)

Example of statistics returned in ordered_map.txt X=1757.0 Y=65_stats.txt (obtained from examples/ex_gui_users_defined_models_2d_map.py):

[[Fit Statistics]]
# fitting method   = leastsq
# function evals   = 93
# data points      = 575
# variables        = 17
chi-square         = 112158.281
reduced chi-square = 201.000503
Akaike info crit   = 3066.14539
Bayesian info crit = 3140.16968
R-squared          = 0.97525302
[[Variables]]
m01_ampli:  914.225148 +/- 10.5474510 (1.15%) (init = 950.0121)
m01_fwhm:   9.56464045 +/- 0.20679661 (2.16%) (init = 9.594399)
m01_x0:     520.183187 +/- 0.05449477 (0.01%) (init = 520.1866)
m02_ampli:  25.7308415 +/- 3.15171955 (12.25%) (init = 44.85308)
m02_fwhm:   89.0443082 +/- 20.7241979 (23.27%) (init = 88.40916)
...
slope:     -0.01339502 +/- 0.00379213 (28.31%) (init = -0.01434605)
intercept:  35.2928118 +/- 2.39248326 (6.78%) (init = 35.87932)
[[Correlations]] (unreported correlations are < 0.100)
C(slope, intercept)     = -0.8110
C(m02_fwhm, intercept)  = -0.6385
C(m04_fwhm, m04_x0)     = +0.6378
C(m01_fwhm, m05_x0)     = +0.5982
C(m01_fwhm, m05_fwhm)   = -0.5258
C(m05_fwhm, m05_x0)     = -0.5191
...

Example of the fitted parameters in ordered_map.txt X=1757.0 Y=65.csv:

Also, a file, named results.csv, that concatenates all the parameters (ordered by names) of all the fitted spectra, is automatically generated when clicking on Save Results in the GUI.

Example of results.csv (extract):

Finally, when dealing with 2d-map spectra, the button Export enables to write the current displayed data in a .csv file as: {spectrum basename}_{parameter name}_{label}.csv file.

Example of ordered_map_ampli_2.csv (extract):