jaxley

library0.13.0pypypi✓ verified 90d ago

Jaxley is a Python library for differentiable neuron simulations built on JAX. Current version is 0.13.0, released with monthly cadence. It enables automatic differentiation through neuron models, supporting multi-compartment simulations, channel dynamics, and synaptic connections.

ai-ml

pip install jaxley

INSTALL

IMPORT

SIG · JAXLEY

jaxley

ai-mlpythonv0.13.0

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

✓ import jaxley as jx

✗ import jaxley

Common practice is to import as jx for brevity, but direct import works. However, many examples use jx.

Module

✓ from jaxley import Module

✗ from jaxley.base import Module

Module class used to be in a private submodule; now it's at the top level.

Quickstart: create a single-compartment neuron, apply a step current, record membrane potential, and run simulation.

import jaxley as jx import jax.numpy as jnp # Create a single compartment cell cell = jx.Module() cell.add('compartment') # Configure stimuli and record cell.stimulate(jx.step_current(amplitude=0.1, duration=10.0)) cell.record() # Simulate for 20 ms voltage = cell.integrate(dt=0.025, t_max=20.0) print(voltage.shape) # (1, 800) - one compartment, 800 time steps

Debug

Known issues

breakingBetween v0.2 and v0.3, the API for creating compartments was overhauled. `jx.Module` replaces the older `jx.single_compartment()`.

fix

Use `jx.Module()` and call `.add('compartment')` instead of `jx.single_compartment()`.

affects: 0.2.x -> 0.3+

breakingThe `integrate()` method now returns JAX arrays, not NumPy arrays. Direct indexing or conversion may be needed.

fix

Convert with `np.asarray(voltage)` if NumPy operations are required.

affects: all

gotchaJaxley requires JAX and defaults to float32. For simulations requiring double precision, you must set JAX configuration: `jax.config.update('jax_enable_x64', True)` before importing jaxley.

fix

Set `jax.config.update('jax_enable_x64', True)` before importing jaxley.

affects: all

Errors

Common errors & fixes

AttributeError: module 'jaxley' has no attribute 'single_compartment'

Removed in v0.3; the API changed to use `Module`.

fix

Use `jx.Module()` and add compartments via `.add('compartment')`.

JaxleyError: Cannot automatically differentiate this channel mechanism. Consider using jaxley.channels instead of custom functions.

Custom ion channel dynamics not supported for autodiff if not implemented as a Jaxley channel.

fix

Rewrite custom channel models using `jx.channels.HHChannel` or other built-in channel classes.

Upgrade

Version history

0.13.0latest on PyPI · released Dec 18, 2025

Audit

Dependencies

jaxrequiredCore backend for automatic differentiation and GPU acceleration.

jaxlibrequiredRequired by JAX for hardware acceleration.

Agent activity

12 hits · last 30 days

node

Resources

No resource links recorded.