Output files#

All output files are written to the output directory defined in the project block of the input YAML file.

For susceptibility fitting workflows (fit_susc), the names of all generated files are printed to the terminal together with a short description. Depending on the selected options and workflow configuration, the following output files may be produced:

assigned_experiment_<TEMPERATURE>_K.csv Assigned experimental data at a given temperature.

Written in the wide-format CSV layout (see Experiment CSV format in Input YAML Files). The file is identical in structure to the input experimental CSV, with signal assignments updated to reflect the result of the fitting or assignment step.
dft_hyperfines.csv Raw hyperfine coupling constants extracted from a DFT calculation.

Generated only when hyperfine: method: dft is selected.
hyperfines_and_shifts_<TEMPERATURE>_K.csv Combined hyperfine and shift data at a given temperature.

Contains hyperfine coupling constants, chemical shifts, atomic coordinates, and chemical labels for each nucleus in the system.
pcs_isosurf_<TEMPERATURE>_K.cube Pseudocontact shift isosurface.

Generated as a Gaussian cube file for visualisation of PCS fields in real space.
susceptibility_components_chi.pdf Temperature dependence of the magnetic susceptibility components.

Shows \(\chi\) versus \(T\). Generated when more than one temperature is specified and --isoaxrh_plots on or --save is enabled.
susceptibility_components_chiT.pdf Temperature-scaled susceptibility components.

Shows \(\chi T\) versus \(T\). Generated when more than one temperature is specified and --isoaxrh_plots on or --save is enabled.
susceptibility_tensor.csv Fitted magnetic susceptibility tensors.

Contains susceptibility tensors as a function of temperature, standard deviations of fitted parameters, goodness-of-fit metrics (\(r^2\), \(r^2_\mathrm{adj}\), MAE), eigenvalues of the susceptibility tensor, and an eigenvector representation using Euler angles.
shift_width_bubble_<TEMPERATURE>_K[_<ISOTOPE>].pdf Scatter plot of observed chemical shift (x) versus linewidth (y) for each signal at a given temperature.

Two panels are shown side by side:
- Matched (left): signals whose assignment is present in both the experimental file and the molecule. Experimental markers are filled and scaled proportionally to the integrated peak area. When an r⁻⁶ fit is available, predicted values are overlaid as open circles at the predicted shift position, connected to the corresponding experimental point by a dashed line.
- Unmatched (right): signals present in only one dataset, drawn in a distinct highlight colour. Experimental unmatched markers are area-scaled (same convention as the matched panel). Predicted unmatched markers (open circles) are scaled by the number of equivalent nuclei sharing that chemical label (group size).
Assignment labels are annotated next to every marker.

Generated when relaxation fitting is enabled or experimental linewidth data are present. A per-isotope suffix is appended to the file name when the molecule contains more than one nuclear isotope.
shift_r1_bubble_<TEMPERATURE>_K[_<ISOTOPE>].pdf Identical layout to shift_width_bubble but with longitudinal relaxation rate R₁ (s⁻¹) on the y-axis.

Generated only when R₁ data are present in the experimental file or an R₁ r⁻⁶ fit has been performed.
r6_fit_width[_<ISOTOPE>]_<TEMPERATURE>_K.pdf / r6_fit_r1[_<ISOTOPE>]_<TEMPERATURE>_K.pdf Observed versus fitted r⁻⁶ distance model scatter plots.

Experimental values (linewidth or R₁) are plotted against r⁻⁶ for each signal. The smooth fitted curve \(p_1 r^{-6} + p_2\) is overlaid. Fitted parameters \(p_1\), \(p_2\), and RMSE are shown in an annotation box.
r6_tau_space_width[_<ISOTOPE>]_<TEMPERATURE>_K.pdf / r6_tau_space_r1[_<ISOTOPE>]_<TEMPERATURE>_K.pdf τ parameter space plot for a single observable and temperature.

Shows the (τ_e, τ_R) plane. The contour where \(p_1^\mathrm{calc} = p_1^\mathrm{fit}\) (exact match) is drawn as a solid line, with dashed lines marking the bootstrap confidence-interval boundary (default 95%). The observable is identified in a legend. When a fixed τ_R is supplied, arrows mark the derived τ_e at the intersection and the τ values are annotated. Both axes are logarithmic and labelled in picoseconds.
r6_tau_space_combined[_<ISOTOPE>]_<TEMPERATURE>_K.pdf Overlay of linewidth and R₁ τ-space constraints on one plot.

The exact-match contour for each observable is drawn as a solid line in a distinct colour (primary palette for R₁, highlight colour for linewidth). Dashed lines of the same colour show the bootstrap confidence interval boundaries. The region where both observables are simultaneously consistent with their fitted values is where the two contours intersect.

Generated only when both linewidth and R₁ fits are available.
r6_tau_space_multitemp_width[_<ISOTOPE>].pdf / r6_tau_space_multitemp_r1[_<ISOTOPE>].pdf Multi-temperature τ-space plot for a single observable.

Each temperature produces one constraint contour (solid line) and a shaded confidence band, coloured on a sequential palette from low to high temperature. The intersection of all contours identifies the (τ_e, τ_R) pair consistent with every experimental temperature simultaneously.

Generated when relaxation data are available at more than one temperature.
peak_data_<TEMPERATURE>_K[_<FIELD>_T].csv Per-chemical-label averaged shifts, linewidths, and relaxation-rate decomposition. Written by both the fit_susc and predict workflows.

The header comment records the temperature T, the static field B0, and — when a relaxation model was evaluated — the correlation times τ_R and τ_e. The magnetic field is also embedded in the file name because the relaxation rates are field-dependent.

Columns (only those with available data are written):
- chem_label, isotope, count — chemical label, its nuclear isotope, and the number of equivalent nuclei sharing the label.
- δ_total_avg, δ_dia_avg, δ_pc_avg (ppm) — total, diamagnetic, and pseudocontact shift components.
- δ_fc_avg (ppm) — Fermi-contact shift. Named δ_fc_g_corr_avg or δ_fc_spin_only_avg when a g-corrected or spin-only isotropic susceptibility is used; Δδ_fc_g_corr_avg gives the g-correction relative to the spin-only reference when both are available.
- δ_orb_avg, δ_orb_iso_avg, δ_orb_aniso_avg (ppm) — orbital shift contributions, written only when an orbital contribution is available.
- linewidth_avg (ppm) — predicted linewidth. Named linewidth_avg_relax when derived from a relaxation model, or linewidth_avg_auto when neither a relaxation model nor an r⁻⁶ linewidth fit is available, in which case a cosmetic display width (a fixed fraction of the shift range) is written so the spectrum remains drawable.
- R1_total, R1_sbm_dipolar, R1_sbm_contact, R1_curie (s⁻¹) — longitudinal relaxation rate and its Solomon–Bloembergen–Morgan dipolar/contact and Curie components.
- R2_sbm_dipolar, R2_sbm_contact, R2_curie (s⁻¹) — transverse relaxation-rate components. The R₂ total is not written separately because it equals linewidth × π|γ|B₀.
The relaxation-rate columns are present only when a relaxation model has been evaluated: predict with a relaxation block, or fit_susc when a τ_R estimate and an r⁻⁶ relaxation fit allow τ_e to be derived.