Global-local fitting across stacked EIS datasets. The module supports all circuit elements listed in Circuit models.
from EISFitpython import singlechiELEMENT_PARAM_COUNT = {
"R": 1, "C": 1, "L": 1, "W": 1,
"Q": 2, "F": 2, "G": 2, "H": 3,
}This mapping aligns structured fit parameters and report labels.
flatten_params(params, circuit_str=None, N_sub=None)Flatten a structured global-local parameter collection for the optimizer.
| Parameter | Description |
|---|---|
params |
Circuit-ordered scalars and iterables. |
circuit_str |
Optional numbered circuit string used to walk parameter slots. |
N_sub |
Number of datasets. Iterables of exactly this length are local. |
Scalars become one value. Iterables of length N_sub contribute every value.
Single-value iterables provide a global starting value. Returns a Python list.
Single_chi(f, *params, circuit_str, circuit_type="fit")Split a stacked frequency sweep, distribute structured parameter values to each dataset, evaluate the circuit, and combine the result.
| Parameter | Description |
|---|---|
f |
Concatenated frequency array. Boundaries are detected by sweep reset. |
*params |
Scalars for global slots; iterables of dataset length for local slots. |
circuit_str |
Circuit expression. Required keyword-only argument. |
circuit_type |
"fit" for concatenated real/imaginary output; "predict" for complex output. |
Returns a real array of length 2 * len(f) in fit mode or a complex array of
length len(f) in prediction mode.
format_circuit_output(
popt,
perror,
CorrM,
circuit_str,
N_sub,
Temp,
param_template,
)Build circuit-aware parameter and correlation tables.
param_template contains one Boolean per circuit parameter slot: True means the
slot is local and consumes N_sub optimized values; False means global and
consumes one. Temp values are inserted into local labels with a °C suffix.
Returns (parameter_df, correlation_df) as pandas DataFrames. The parameter table
contains Fit_Params, Value, Error, and % error. The correlation table is
limited to the number of generated labels and masks its diagonal and upper triangle.
Single_chi_report(f, Z, params, Temp, circuit_str, weight_mtd="M")Run the complete global-local fit, format and print statistics, write per-temperature fitted data, generate plots, and write a report.
| Parameter | Description |
|---|---|
f, Z |
Concatenated frequency and complex-impedance arrays. |
params |
Structured circuit parameters using scalar/global and iterable/local semantics. |
Temp |
Temperatures in Celsius; one per detected dataset. |
circuit_str |
Numbered circuit expression. |
weight_mtd |
"M", "P", or "U". |
Returns (popt, perror, Z_fit). The optimization uses method="lm".
The function writes Single-chi_Report.txt, S-chi_fit_data_<Temp>C.txt for each
dataset, individual Nyquist SVGs, and one stacked Bode SVG.
generate_plots(f, Z, Z_fit, Temp)Split stacked data by sweep reset, create one Nyquist plot per paired temperature, create a stacked Bode plot, save the SVG files, and display all figures.
Returns (last_nyquist_figure, bode_figure). Supply one temperature value for each
dataset.