API Reference

Canonical import:

import jaxfne as jtfne

This page is the complete index of the public API (jaxfne.__all__, 195 names), grouped by module. Per-module pages carry detailed signatures and examples. Current release jaxfne==0.3.42 (tag v0.3.42). The root-level export helpers introduced in the v0.3.37/v0.3.38 line remain formal __all__ members.

Scope & truth gates

All field/EEG/MEG/EMM outputs are computational proxies (claim_level = "computational_scaffold", field_solver_status = "linear_solver", field_claim_level = "proxy_readout", physical_amplitude_calibrated = False). See Limitations and future plans for the scope statement.

Module pages

Page	Covers	Public names
Core	Configuration, Model, Simulation, Signals, readouts, receipts, suites	69
Emitters	Izhikevich emitter, receptors, synapses, EIG networks, edge lists	19
Fields	Source→laminar projection, FieldOutput, proxy diagnostics	12
Probes	EEG/MEG/EMM proxy transforms (within Fields)	(in Fields)
Objectives	Objective, ObjectiveReport, rate targets	(in Core)
Runtime	RuntimeConfig, enable_x64, runtime_report	(in Core)
Validation	config/field validators, operator_status, is_valid_signal	2
(no page yet)	Optimizers (optim, 15) · IO/receipts (10) · Export & figures (6) · Bridges (7) · Paradigms (6) · Solvers (7) · Sanity-delta (7) · Plasticity (5) · Tutorial utils (4) · Sharding (4) · Connectivity (3) · PyNWB (2) · Experimental HPC (2) · JAX Spectral Analysis (6) · geometry/builders/streaming/stimulus (4)	93

Several public names (optimizers, IO, export, bridges, paradigms, sharding, solvers) do not yet have a dedicated module page — they are listed with full signatures in the complete symbol index below. Counts sum to 195 (len(jaxfne.__all__)). See the docs audit (internal_docs/docs_audit_v0330.md) for the page-migration plan.

Minimal workflow (verified)

import jaxfne as jtfne

cfg = jtfne.suite2_four_celltype_config(seed=0, duration_ms=1000.0, dt_ms=0.1)
model = jtfne.construct(cfg)
signals = jtfne.simulate(model, duration_ms=1000.0, dt_ms=0.1, seed=0)

vm = signals.get("vm")            # membrane voltage [T, N]
spk = signals.get("spk")          # spikes [T, N]
e_idx = model.select(cell_type="E")     # excitatory neuron indices
vm_e = signals.get("vm", cell_type="E") # equivalent to vm[:, e_idx]
assert vm_e.shape[-1] == len(e_idx)

---

Export & figure APIs (root-level)

The strict-notebook export grammar introduced in jaxfne==0.3.37 is exposed as root-level callables (jaxfne.<name>) and, as of v0.3.38, is registered in jaxfne.__all__ as formal public API (see the Export & figures group in the symbol index below). These are the canonical replacement for direct matplotlib/json calls in release-facing notebooks. matplotlib is imported lazily inside the plotting/save functions, so importing jaxfne does not pull in a plotting backend.

Symbol	Kind	Summary
save_figure	func	Save a matplotlib figure to disk.
save_figures	func	Save multiple figures to an output directory.
export_report	func	Export a complete report with JSON artifacts and figures.
export_tutorial_artifacts	func	Export tutorial artifacts (JSON only, no figures).
plot_raster	func	Plot a spike raster.
plot_spectrolaminar_suite	func	Plot spectrolaminar suite from a signals object.

All export helpers honor the truth gates: JSON is written with allow_nan=False, and figure/readout outputs remain proxy diagnostics (physical_amplitude_calibrated = False).

---

Complete public symbol index

func/class/const/module as resolved from jaxfne.__all__ (195 names, grouped by defining module). Summaries are the first docstring line; _(undocumented)_ marks public callables with no docstring in the released wheel.

Core (69)

Symbol	Kind	Summary
AxisSpec	class	Typed descriptor for one tensor axis in the TFNE scaffold.
BasisSpec	class	Typed descriptor for the computation basis of a TFNE run.
Config	class	Declarative TFNE model configuration.
config_to_configuration	func	Map the network/emitter/field/probes sections to a Configuration.
config_to_geometry	func	Map the geometry section to a LaminarSourceGeometry, or None.
config_to_simulation	func	Map the run section of a JaxFNEConfig to a Simulation.
config_to_trial_batch	func	Map the trials section and conditions to a TrialBatch.
config_truth_boundary	func	Return a JSON-safe copy of the truth boundary section.
Configuration	class	Declarative TFNE model configuration.
configuration	func	— (undocumented)
ConfigValidationResult	class	Report container for configuration validation.
construct	func	— (undocumented)
dataset_spec	func	Return a DatasetSpec schema declaration.
DatasetSpec	class	Manifest-safe dataset/comparison declaration for observed data.
default_basis_spec	func	Return the default BasisSpec matching the current laminar-proxy scaffold.
enable_x64	func	Enable JAX float64 mode before constructing arrays and report status.
get_signal	func	Thin free-function accessor that delegates to Signals.get.
JaxFNEConfig	class	JSON-safe container for a complete .jcfg.json TFNE specification.
laminar_source_geometry	func	Build a LaminarSourceGeometry from an ordered population sequence.
LaminarPopulation	class	Metadata descriptor for one named laminar cell population.
LaminarSourceGeometry	class	Metadata descriptor for the full laminar source geometry.
load_config	func	Load a .jcfg.json file and return a JaxFNEConfig.
matrix_parameter	func	Create a matrix parameter specification for tuning weight matrices.
MatrixParameterSpec	class	Declarative specification for a tunable weight matrix parameter.
migrate_schema	func	Upgrade a legacy truth/metadata dict to the canonical truth-gate schema.
Model	class	— (dataclass; fields in signature)
Net	class	— (dataclass; fields in signature)
Objective	class	Declarative objective specification: losses, regularizers, and diagnostic gates.
objective	func	— (undocumented)
ObjectiveReport	class	Structured, immutable result of evaluating an Objective against Signals.
operator_status	func	Return the current operator status registry for all declared operators.
Probe	class	— (dataclass; fields in signature)
provenance_receipt	func	Capture release provenance atomically.
rate_targets	func	Create a multi-group firing-rate objective.
readout_spec	func	Build a ReadoutSpec for declarative feature extraction.
ReadoutResult	class	Result of applying a ReadoutSpec to Signals.
ReadoutSpec	class	Declarative specification for extracting a scalar feature from Signals.
run_receipt	func	Build a RunReceipt for a completed simulation run.
run_trials	func	Execute a batch of trials using the model.
RunReceipt	class	Complete, JSON-safe record of a single simulation run.
runtime	func	— (undocumented)
runtime_report	func	— (undocumented)
RuntimeConfig	class	JAX runtime and dtype policy.
Signal	class	Simulation output container holding multiple arrays.
Signals	class	Simulation output container holding multiple arrays.
simulate	func	Run a simulation with the given model.
Simulation	class	— (dataclass; fields in signature)
simulation	func	— (undocumented)
standard_visual_omission	func	Construct a Paradigm with standard visual oddball/omission task conditions.
stimulus_schedule	func	Build a StimulusSchedule from a sequence of events.
StimulusSchedule	class	Explicit drive schedule for event-aligned stimulus injection.
suite2_celltype_presets	func	Return compact E/PV/SST/VIP reduced-emitter preset metadata.
suite2_four_celltype_config	func	Build the Suite No. 2 four-emitter E/PV/SST/VIP configuration.
suite2_net1_config	func	Build net1: a uniformly sampled 3D E/PV/SST/VIP column.
suite2_run_bundle	func	Run simulation, readouts, manifest, and receipt for Suite No. 2 notebooks.
suite2_simulation	func	Create a Suite No. 2 simulation with deterministic runtime metadata.
suite2_single_neuron_config	func	Build the Suite No. 2 one-emitter configuration.
suite2_tune_noise_agsdr_adam	func	Tune Poisson-drive amplitude toward a target mean firing-rate range.
suite2_v1_v4_config	func	Build the Suite No. 2 V1-V4 laminar scaffold with six layers per area.
surrogate_config	func	Return a SurrogateConfig declaration for an Optax gradient path.
SurrogateConfig	class	Declared surrogate-gradient metadata for discontinuous emitter paths.
trial_batch	func	Create a TrialBatch by repeating conditions.
TrialBatch	class	A collection of trial specifications to be run.
TrialBatchResult	class	Results from a batch of trials.
TrialResult	class	Result of a single simulation trial.
TrialSpec	class	Specification for a single simulation trial.
TuneResult	class	Result object returned by Model.tune() with multi-parameter optimization.
validate_config	func	Validate a JaxFNEConfig and return a ConfigValidationResult.
with_emitter_parameters	func	Functional wrapper for Model.with_emitter_parameters.

Emitters (19)

Symbol	Kind	Summary
EdgeList	class	Sparse recurrent connectivity as a JAX pytree.
EIGNetwork	class	Lightweight description of an E/PV/SST/VIP-like reduced network.
Emitter	class	Base class for package-level emitter facades.
GLIFEmitter	class	Base class for package-level emitter facades.
izhikevich_params_from_labels	func	Create reduced Izhikevich parameters from explicit cell labels.
IzhikevichEmitter	class	Reduced Izhikevich emitter facade with a JAX step function.
IzhikevichParams	class	Parameter container for a reduced Izhikevich population.
LIFEmitter	class	Base class for package-level emitter facades.
make_edge_list_from_dense	func	Convert a dense recurrent weight matrix into a sparse EdgeList.
make_eig_network	func	Build a minimal EIG network with laminar depth positions.
ReceptorSpec	class	Metadata declaration for a synaptic receptor. Not a biological kernel.
simulate_edge_recurrent_izhikevich	func	Simulate reduced Izhikevich emitters with sparse recurrent synapses.
simulate_eig_izhikevich	func	Simulate a reduced EIG Izhikevich scaffold using jax.lax.scan.
simulate_receptor_exponential_izhikevich	func	v0.0.11 receptor-indexed exponential recurrent kernel.
standard_receptor_specs	func	Provide standard declarative receptor metadata. No biological claim.
standard_receptor_tau_table	func	Return the receptor_index → tau_ms lookup table used by v0.0.11.
SynapseLayer	class	Exponential synapse layer returning recurrent input currents.
SynapseSpec	class	Metadata declaration for a synapse. Not a biological kernel.
SynapseState	class	— (no docstring)

Fields (12)

Symbol	Kind	Summary
compute_conservation_proxy_diagnostics	func	Compute conservation-inspired proxy diagnostics over existing source/field arrays.
construct_source_tensor	func	— (undocumented)
eeg_proxy_transform	func	Compute EEG-proxy readout via linear leadfield projection.
emm_proxy_transform	func	Compute EMM-proxy (normalized activity/source/field cost) readout.
FieldOutput	class	Container for laminar proxy field/readout arrays.
LinearReadout	class	— (dataclass; fields in signature)
meg_proxy_transform	func	Compute MEG-proxy readout via linear leadfield projection.
probe_laminar_modes	func	— (undocumented)
project_laminar_sources	func	Project source traces to laminar proxy contacts.
project_sources_to_laminar_field	func	— (undocumented)
validate_projection_invariants	func	— (undocumented)
validate_source_field_status	func	Return truth-preserving status for source-field readouts.

Optimizers — optim (15)

Symbol	Kind	Summary
AGSDR	class	Legacy AGSDR adapter retained for old notebooks and tests.
agsdr	func	Return an optimizer spec for AGSDR.
agsdr_transform	func	Return an Optax-compatible GradientTransformation for Adaptive GSDR.
AGSDROptimizerSpec	class	Multi-parameter AGSDR optimizer specification with execution parameters.
AGSDRState	class	Adaptive Genetic Stochastic Delta Rule optimizer state.
gsdr	func	Return an OptimizerSpec for the GSDR (Genetic Stochastic Delta Rule) optimizer.
gsdr_transform	func	Return an Optax-compatible GradientTransformation for Genetic SDR.
GSDRState	class	Genetic Stochastic Delta Rule optimizer state.
optax_adam	func	Return an OptimizerSpec for Optax Adam.
optax_sgd	func	Return an OptimizerSpec for Optax SGD.
OptimizerSpec	class	Declarative optimizer specification with differentiability metadata.
random_search	func	Return an OptimizerSpec for random search.
require_optax	func	Import Optax lazily with an informative error.
sdr_transform	func	Return an Optax-compatible GradientTransformation for Stochastic Delta Rule.
SDRState	class	Stochastic Delta Rule optimizer state.

IO & receipts (10)

Symbol	Kind	Summary
asset_hashes	func	Create a SHA256 hash manifest for assets.
config_hash	func	Return a compact SHA256 hash for a configuration-like object.
json_safe	func	Convert common scientific Python/JAX objects into strict JSON values.
manifest	func	Build a strict JSON-safe run manifest.
probe_report	func	Create a probe operator report JSON bundle.
save_json	func	Save strict JSON with allow_nan=False.
save_receipt	func	Save a RunReceipt as strict JSON.
sha256_file	func	Return SHA256 for a file.
sha256_text	func	Return SHA256 for a text payload.
validation_report	func	Create a validation report JSON bundle.

Export & figures (6)

Symbol	Kind	Summary
export_report	func	Export a complete report with JSON artifacts and figures.
export_tutorial_artifacts	func	Export tutorial artifacts (JSON only, no figures).
plot_raster	func	Plot a spike raster.
plot_spectrolaminar_suite	func	Plot spectrolaminar suite from signals object.
save_figure	func	Save a matplotlib figure to disk.
save_figures	func	Save multiple figures to an output directory.

Bridges — Jaxley (7)

Symbol	Kind	Summary
BridgeSpec	class	JSON-safe optional-backend bridge declaration.
hh_numpy_reference_trace	func	Standalone tutorial/reference Hodgkin-Huxley single-compartment trace.
jaxley_trace_to_signals	func	Convert Jaxley-style voltage trace array to jaxfne Signals.
JaxleyBridge	class	Jaxley-focused biophysical emitter bridge.
JaxleyEmitterBridge	class	Jaxley bridge contract for reserved compartment emitters.
JaxleyTraceSpec	class	Metadata specification for Jaxley-style voltage trace arrays.
require_jaxley	func	Import Jaxley lazily with an informative error.

Paradigms (6)

Symbol	Kind	Summary
coop_omission_oddball_paradigm	func	Create a Continuous Omission Oddball Paradigm (COOP) stimulus sequence.
omission_oddball_paradigm	func	Create an omission/oddball detection paradigm.
Paradigm	class	— (dataclass; fields in signature)
paradigm	module	(constant; see source)
ParadigmCondition	class	A specific trial condition: sequence of stimuli and associated events.
ParadigmEvent	class	Discrete event within a task trial: stimulus, behavioral code, or omission marker.

Solvers (7)

Symbol	Kind	Summary
DiffraxSolver	class	Optional Runge-Kutta solver using diffrax (lazily imported).
euler_scan	func	Forward Euler integration scan (backward compatibility).
euler_step	func	Single forward Euler step (backward compatibility).
EulerSolver	class	Forward Euler integrator using JAX and lax.scan.
solve_ode	func	Public ODE solver entrypoint routing to appropriate solver backend.
solve_volume_conductor_experimental	func	Experimental volume conductor solver skeleton.
SolverConfig	class	Configuration class for ODE solvers.

Sanity-delta runtime (7)

Symbol	Kind	Summary
BackupState	class	Resumable task state with ring buffer history.
BehaviorGate	class	Fixation gate: monitors PFC superficial activity.
HierarchicalOddballParadigm	class	Task paradigm: AAAB oddball sequence with timing and gating.
Manifest	class	Output manifest: configuration, paradigm, backup, validation.
SanityDeltaConfig	class	Hierarchical oddball configuration factory and validation.
SanityDeltaModel	class	Wrapper around constructed hierarchical oddball model.
TaskEpisode	class	Result of a task episode with probing, export, validation.

Plasticity (5)

Symbol	Kind	Summary
plot_stdp_adaptation_suite	func	Generates and saves the standard STDP adaptation visualization figures.
STDPPlasticityConfig	class	Configuration class for STDP activity-dependent plasticity.
STDPState	class	Container for the state variables of the STDP synapse model.
summarize_stdp_adaptation	func	Computes synapse-by-synapse adaptation statistics.
update_stdp_weights_jax	func	JAX-optimized plasticity weight update kernel (STDP).

Tutorial utils (4)

Symbol	Kind	Summary
build_tutorial_laminar_column	func	Build a laminar column scaffold model.
kappa_synchrony	func	Compute spike synchrony measure (kappa statistic) across neurons.
rate_synchrony_targets	func	Create an objective specification for AGSDR tuning toward rate and synchrony targets.
select_neurons	func	Select neuron indices matching given criteria (area, layer, cell_type).

Sharding (4)

Symbol	Kind	Summary
get_sharding_context	func	Return a dict with mesh, candidate, and replicated sharding specs.
make_candidate_sharding	func	Return a jax.sharding.NamedSharding that slices the first
make_population_mesh	func	Return a 1-D named jax.sharding.Mesh across all visible JAX devices.
make_replicated_sharding	func	Return a jax.sharding.NamedSharding that fully replicates an array

Connectivity (3)

Symbol	Kind	Summary
compile_connection_rules	func	Compile declared connection rules into sparse finite edge arrays.
compile_connection_rules_jax	func	Tensorized JAX connectivity compiler producing static-shape edge outputs.
ConnectionCompileResult	class	Compiled sparse connectivity.

Geometry (1)

Symbol	Kind	Summary
make_ei_cloud_network	func	Generates geometry and initial weights for a 100-neuron E-I cloud network.

Builders (1)

Symbol	Kind	Summary
laminar_cortex_config	func	Generalized laminar cortical configuration builder.

Streaming (1)

Symbol	Kind	Summary
run_stdp_stream	func	Runs simulation in a chunked, streaming fashion to avoid memory explosion.

Stimulus (1)

Symbol	Kind	Summary
triangular_drive	func	Generates a triangular drive trace.

JAX Spectral Analysis (6)

Symbol	Kind	Summary
spectrolaminar_psd_jax	func	Compute spectrolaminar PSD averaged across trials using JAX.
bandpower_jax	func	Compute average power within a frequency band normalized by channel max.
spectrolaminar_readout_kernel_jax	func	Batchable readout kernel computing relative power and normalized band profiles.
spectrolaminar_similarity_kernel_jax	func	Compute the profile similarity score in JAX.
spectrolaminar_similarity_candidates_jax	func	Batched vectorization path for similarity scoring.
spectrolaminar_similarity_candidates_seeds_jax	func	Nested batched vectorization path for seeds and candidates.

Validation registry (2)

Symbol	Kind	Summary
compilation_registry	const	Automated JAX tracing and compilation tracking registry.
is_valid_signal	func	Check if signal arrays contain only finite values (no NaN/Inf).

PyNWB compatibility (2)

Symbol	Kind	Summary
read_nwb	func	Placeholder for NWB read (reserved status).
write_nwb	func	Placeholder for NWB write (reserved status).

Experimental HPC (2)

Symbol	Kind	Summary
NodeIdentity	class	Stable node identity for selector-addressable circuits.
SelectorSpec	class	Selector over area/layer/cell-type/id fields.

Submodules (1)

Symbol	Kind	Summary
vis	module	Visualization package for jaxfne.

Constants (4)

Symbol	Kind	Summary
_KNOWN_METRICS	const	⚠ private name leaking into __all__ — see docs audit (remove).
CELL_TYPE_PRESETS	const	Mapping of cell-type label → preset Izhikevich parameters.
DEFAULT_SPIKE_IMPULSE_GAIN	const	Default spike-impulse gain for the source proxy.
RECEPTOR_KINETICS	const	Mapping of receptor name → kinetic time constants.

---

See the docs audit & restructure notes (internal_docs/docs_audit_v0330.md) for orphaned pages, duplicate cleanup, and the per-module table migration.