Home
Jaxley
is a differentiable simulator for biophysical neuron models in JAX. Its key features are:
- automatic differentiation, allowing gradient-based optimization of thousands of parameters
- support for CPU, GPU, or TPU without any changes to the code
jit
-compilation, making it as fast as other packages while being fully written in python- backward-Euler solver for stable numerical solution of multicompartment neurons
- elegant mechanisms for parameter sharing
Getting started¶
Jaxley
allows to simulate biophysical neuron models on CPU, GPU, or TPU:
import matplotlib.pyplot as plt
from jax import config
import jaxley as jx
from jaxley.channels import HH
config.update("jax_platform_name", "cpu") # Or "gpu" / "tpu".
cell = jx.Cell() # Define cell.
cell.insert(HH()) # Insert channels.
current = jx.step_current(i_delay=1.0, i_dur=1.0, i_amp=0.1, delta_t=0.025, t_max=10.0)
cell.stimulate(current) # Stimulate with step current.
cell.record("v") # Record voltage.
v = jx.integrate(cell) # Run simulation.
plt.plot(v.T) # Plot voltage trace.
If you want to learn more, we have tutorials on how to:
- simulate morphologically detailed neurons
- simulate networks of such neurons
- set parameters of cells and networks
- speed up simulations with GPUs and jit
- define your own channels and synapses
- define groups
- read and handle SWC files
- compute the gradient and train biophysical models
Installation¶
Jaxley
is available on pypi
:
pip install jaxley
Jaxley
with CPU support. If you want GPU support, follow the instructions on the JAX
github repository to install JAX
with GPU support (in addition to installing Jaxley
). For example, for NVIDIA GPUs, run
pip install -U "jax[cuda12]"
Feedback and Contributions¶
We welcome any feedback on how Jaxley
is working for your neuron models and are happy to receive bug reports, pull requests and other feedback (see contribute). We wish to maintain a positive community, please read our Code of Conduct.
License¶
Apache License Version 2.0 (Apache-2.0)
Citation¶
If you use Jaxley
, consider citing the corresponding paper:
@article{deistler2024differentiable,
doi = {10.1101/2024.08.21.608979},
year = {2024},
publisher = {Cold Spring Harbor Laboratory},
author = {Deistler, Michael and Kadhim, Kyra L. and Pals, Matthijs and Beck, Jonas and Huang, Ziwei and Gloeckler, Manuel and Lappalainen, Janne K. and Schr{\"o}der, Cornelius and Berens, Philipp and Gon{\c c}alves, Pedro J. and Macke, Jakob H.},
title = {Differentiable simulation enables large-scale training of detailed biophysical models of neural dynamics},
journal = {bioRxiv}
}