jajcayn / interactive-spiking-neuron-model

Interactive spiking neuron models

Interactive spiking neuron models

Interactive spiking neuron models

Interactive models of spiking neuron in jupyter notebook! You can play with model (intrinsic) and integration parameters (external current and integration length) and see how the dynamics of the neuron evolves.

Launch interactive binder

Nothing required, just click on the button, wait a bit till binder builds all that is necessary, choose a model notebook and then click Cell -> Run All. The last cell contains the interactive plot.

Homemade

git clone https://github.com/jajcayn/interactive-spiking-neuron-model.git
cd interactive-spiking-neuron-model
pip install -r requirements.txt
jupyter notebook

and launch *_model.ipynb. Then just Cell -> Run All and you are done.

Models

• Izhikevich model
• (others soon)

Implement your own

If you have an idea for outstanding spiking neuron model (or just want to implement an existing one) nothing is easier than do it yourself! Check out models.py how e.g. IzhikevichModel is implemented. In general, all models should be subclassed from model_base.BaseSpikingModel and the following functions and attributes should be reimplemented:

• y0 for initial conditions
• index_voltage_variable for index of membrane potential within whole arbitrarily long state vector
• spike_condition for setting the condition for spike
• required_params for list of required parameters of the model
• _apply_reset(self, y) is a function for applying after-spike reset acting on state vector y
• _rhs(self, t, y, I) is a function that computes single dt step of the model dynamics, i.e. the right hand side of model ODE

And you're done! All other stuff, mainly the integration itself is handled by the BaseSpikingModel. If you are really eager to see interactive demo of some model not yet implemented here but you're no good with python, just drop me a message or open an issue and I'll implement it.

Happy modelling!