コンテンツにスキップ

Chapter 12 Appendix

12.1 Supported Versions

Item Scope
Manual version 0.17
Supported app AutoStatSpectra 0.17.x
Language English
Distribution Web version, Desktop version

This manual covers the AutoStatSpectra 0.17 series. If the screens or operations differ from this manual, check the version of your app.

12.2 Parameters and Terminology

Term Description Related chapter
Dataset A single loaded spectrum data item Chapter 4
Group analysis Analysis that applies the same prior to multiple spectra while keeping results per data item Chapter 9
Prior The parameter range or fixed value specified before Bayesian estimation Chapter 5
Model search range The lower and upper bounds of the number of peaks to evaluate Chapter 5
Peak Function The peak function type, chosen from Gaussian, Lorentzian, and pseudo-Voigt Chapter 5
Center Peak center position. Same unit as the X axis Chapter 5
Height The peak height estimated directly in Height mode Chapter 5
Area (\(S\)) The integrated area estimated directly in Area mode Chapter 5
FWHM The full width at half maximum of a peak Chapter 5
\(\eta_H\) The Gaussian component fraction of pseudo-Voigt in Height mode Chapter 5
\(\eta_A\) The Gaussian component fraction of pseudo-Voigt in Area mode Chapter 5
Set per peak A mode that sets an independent prior for each peak Chapter 5
Offset The background subtracted from the spectrum Chapter 6
Noise model The observation noise model, chosen from Gaussian, Modified Gaussian, and Poisson Chapter 7
Total samples The total number of samples generated during analysis Chapter 8
Burn-in steps In the CPU version, the initial iterations discarded before posterior distribution collection Chapter 8
Params Source The spectrum used as the copy source of the shared conditions in group analysis Chapter 9
Free Energy The Bayesian free energy used for model comparison. Smaller means a better model Chapter 10
Posterior Probability The posterior probability of each model. Between 0 and 1, summing to 1 across the evaluated models Chapter 10
MAP The mode of the posterior distribution (Maximum A Posteriori) Chapter 10
95% Credible Interval The range from the 2.5% to the 97.5% percentile of the posterior distribution Chapter 10

12.3 Differences Between the Web Version and the Desktop Version

Item Web version Desktop version
Runtime environment Browser Desktop app
Analysis method GPU version CPU version
Where analysis runs GPU backend Local PC
Usage management Account, billing, GPU analysis time License key
Network Required Required for license verification
Sampling settings Total samples Total samples, Burn-in steps, Number of Replicas
Error Function Trace Not displayed Displayed
Exchange Rate Warning Not displayed Displayed when the condition applies
energy_trace.csv Not output Output
Advanced Export destination selection Select folder available in Chrome / Edge Select folder available

In the Web version, the analysis time varies with the load on the GPU backend, the number of data points, the number of peaks, and the number of samples. In the desktop version, it depends on the CPU performance of the local PC. For details, see Chapter 1, Chapter 8, Chapter 10, and Chapter 11.

12.4 Supported Formats and Limitations

Loading Analysis Data

Format Extension Loading method
CSV .csv Select rows and columns in Import Settings
TXT .txt Select rows and columns in Import Settings
VAMAS .vms / .npl Loaded directly with a dedicated parser
PHI SOFH .spe / .pro / .ang Loaded directly with a dedicated parser
Excel .xlsx Loaded directly with a dedicated parser

For CSV / TXT, UTF-8 is tried first; if the text appears garbled, the file is read as Shift_JIS / CP932. Import Settings is not available for VAMAS, PHI SOFH, and Excel. For details, including how multiple Y columns are handled, see Chapter 4.

Loading a Reference Offset

The Reference offset can load .csv / .txt / .dat / .tsv / .xlsx / .xls. The reference data must contain the same X coordinates in the same order as the analysis data. See Chapter 6 for details.

Result Output

Purpose Regular output
Numerical data Results ZIP containing CSV files, etc.
Plot images Images ZIP containing PNG files
Reports PDF / Word
Advanced Export Report: PDF / HTML / Markdown, Chart: PNG / SVG / JPG / TIFF / BMP

Note the following limitations.

  • Free Energy and Model Posterior Probability cannot be compared across Height / Area. Compare only within a single run using the same intensity parametrization
  • Group analysis analyzes each spectrum independently; it is not a feature that outputs an averaged spectrum or a combined fit
  • experiment_config.json does not include the peak-count search range, per-peak priors from Set per peak, or fixed Noise values
  • Chart DPI may not be applied to the batch export of some result plots
  • The GPU version does not display the Error Function Trace and does not output energy_trace.csv

For the structure of the output files and additional limitations, see Chapter 11.

12.5 Support and Contact

For inquiries about AutoStatSpectra, please use the contact form on the a.s.ist official website.

When contacting us, please provide the version of the app you are using, whether it is the Web version or the desktop version, the operation during which the problem occurred, and any messages displayed on screen.

12.6 Index

Item Reference
Advanced Export Chapter 11 Exporting Results
Analyze / Analyze all Chapter 8 Sampling Settings and Running Analysis, Chapter 9 Group Analysis
Area ratio constraints Chapter 5 Peak Settings
Data tab / X and Y Range Chapter 4 Data Loading
Error Function Trace Chapter 10 Reviewing Results
Files sidebar Chapter 3 Screen Layout
Free Energy / Posterior Probability Chapter 10 Reviewing Results
Import Settings Chapter 4 Data Loading
Noise tab Chapter 7 Noise Settings
Offset tab / Reference / arPLS Chapter 6 Background Settings
Peak tab / Peak Priors Chapter 5 Peak Settings
Plot Settings Chapter 3 Screen Layout
Sampling tab Chapter 8 Sampling Settings and Running Analysis
Set per peak Chapter 5 Peak Settings
Web version / desktop version Chapter 1 Overview, Chapter 2 Setup
Group analysis Chapter 9 Group Analysis
Supported file formats Chapter 4 Data Loading
How to read results Chapter 10 Reviewing Results