API Reference

This page is auto-generated from the package source by Sphinx’s autosummary extension. Adding a name to nstat.__all__ automatically makes it appear here on the next docs build — there is no manual list to maintain.

The grouping below mirrors the categories in Class Definitions and the public-API section of AGENT_GUIDE.md. Click any symbol to jump to its rendered docstring (with parameter tables, type hints, and links to source).

Core data primitives

SignalObj(time, data[, name, xlabelval, ...])	Multi-dimensional time-series signal object (Matlab SignalObj).
Signal(time, data[, name, xlabelval, ...])	Canonical Pythonic signal abstraction for nSTAT.
Covariate(*args, **kwargs)	Canonical covariate type for model design matrices.
nspikeTrain(spikeTimes[, name, sampleRate, ...])	Point-process (spike train) object (Matlab nspikeTrain).
SpikeTrain(spikeTimes[, name, sampleRate, ...])	Canonical spike train type for point-process analyses.
Events(eventTimes[, eventLabels, eventColor])	Experimental event markers for highlighting epochs in figures.
ConfidenceInterval(time, bounds, *args[, ...])	Confidence interval for a time series or Covariate.

Collections

nstColl([trains])	MATLAB-facing spike-train collection class.
SpikeTrainCollection([trains])	Collection of aligned spike trains for ensemble analyses.
CovColl
CovariateCollection([covariates])	Ordered collection of Covariate objects (Matlab CovColl).
ConfigColl
ConfigCollection([configs])	Ordered collection of TrialConfig objects.

Experiment and configuration

Trial([spike_collection, ...])	Single-trial data container binding spikes, covariates, and events (Matlab Trial).
TrialConfig([covMask, sampleRate, history, ...])	Description of a single GLM fit configuration.
History(windowTimes[, minTime, maxTime, name])	Spike-history basis defined by a set of window boundaries.
HistoryBasis

Modeling and inference

Analysis()	Collection of static methods for GLM analysis of point-process data.
CIF([beta, Xnames, stimNames, fitType, ...])	Conditional Intensity Function for point-process modelling.
CIFModel(time, rate_hz[, name])	Conditional intensity function abstraction used by standalone workflows.
LinearCIF(beta, Xnames, stimNames[, ...])	Conditional intensity function with closed-form derivatives.
FitResult(neuralSpikeTrain, *args, **kwargs)	GLM fit results for one neuron across one or more model configs (Matlab FitResult).
FitSummary(fit_results)	Population-level summary across multiple neurons (Matlab FitResSummary).
FitResSummary(fit_results)	MATLAB-compatible alias for FitSummary.
PopulationTimeRescaleResult(ground_uniforms, ...)	Population-level (multivariate) time-rescaling GOF result.
population_time_rescale(counts_list, ...[, ...])	Multivariate (marked) time-rescaling goodness-of-fit for a population.

psth(spike_trains, bin_edges)

Compute peri-stimulus time histogram (PSTH) from multiple spike trains.

Decoding

DecodingAlgorithms()	Static-method library for neural decoding and state-space estimation.
DecoderSuite()	Canonical decoding API for the Python nSTAT package.
PoissonGLMResult(intercept, coefficients, ...)

fit_poisson_glm(x, y, *[, offset, ...])

Simulation

PointProcessSimulation(time, rate_hz, spikes)
NetworkSimulationResult(time, latent_drive, ...)

simulate_poisson_from_rate(time, rate_hz[, rng])
simulate_cif_from_stimulus(time, stimulus, ...)	Simulate a spike train from a log-linear CIF driven by a stimulus.
simulate_point_process(time, rate_hz, *[, ...])
simulate_two_neuron_network([duration_s, ...])	Standalone Python replacement for the MATLAB/Simulink 2-neuron NetworkTutorial.
run_full_paper_examples(repo_root)

Plot style

set_plot_style([style])	Persist the plotting style for future sessions.
get_plot_style([default])	Return the persisted global plotting style.
apply_plot_style([target, style])	Apply the current nSTAT plot style to a matplotlib figure or axes.

Installation and data

nstat_install(*[, rebuild_doc_search, ...])	Configure the Python package and optionally install example data.
nSTAT_Install(**kwargs)	MATLAB-style alias that delegates to nstat.install.nstat_install().
get_dataset_path(name)
list_datasets()
verify_checksums()
getPaperDataDirs(*[, download])	MATLAB-style tuple-returning alias for get_paper_data_dirs().
get_paper_data_dirs(*[, download])	Return the canonical paper-example data directories.

MATLAB bridge (optional)

is_matlab_available()	Return True if matlab.engine can be imported.
get_matlab_nstat_path()	Resolve the path to the MATLAB nSTAT repo containing .slx models.
set_matlab_nstat_path(path)	Programmatically point to the MATLAB nSTAT repo.

Exceptions and warnings

DataNotFoundError	Raised when a required dataset is missing from the local checkout.
MatlabEngineError	Raised when MATLAB Engine interaction fails.
MatlabFallbackWarning	Issued when MATLAB/Simulink is unavailable and the native Python simulation is used instead.
ParityValidationError	Raised when MATLAB/Python parity validation fails.
UnsupportedWorkflowError	Raised when a legacy workflow has not yet been ported.