{ "metadata": { "name": "slides" }, "nbformat": 3, "nbformat_minor": 0, "worksheets": [ { "cells": [ { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "Brian Model Fitting Tutorial\n", "============================\n", "### CNS 2013\n", "#### Cyrille Rossant" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Spiking neuron models can accurately reproduce the response to a somatically injected current\n", "\n", "![](slides_files/gerstner.PNG)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## How to estimate the parameters?\n", "\n", " * Model-specific optimization algorithms\n", " * Generic optimization with Brian" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## The BRIAN model fitting toolbox\n", "\n", " * Give the equations in mathematical form as usual\n", " * Specify the parameters to optimize\n", " * Specify the injected current and the output spikes\n", " * Run the procedure and get the optimized parameters\n", " \n", "
" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "How it works\n", "============\n", "\n", " * The model is simulated with Brian: one neuron = one parameter set\n", " * The best parameters are found with a stochastic global optimization algorithm (e.g. CMA-ES)\n", " \n", "" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "Getting started\n", "===============\n", "\n", " * You need Brian (it includes directly the model fitting library).\n", " * You also need Playdoh (distributed optimization) ([github.com/rossant/playdoh](https://github.com/rossant/playdoh)):\n", " \n", " pip install playdoh" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Toy example\n", "\n", " * Let's first simulate a simple integrate-and-fire neuron responding to an injected fluctuating current." ] }, { "cell_type": "code", "collapsed": false, "input": [ "%pylab inline" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "\n", "Welcome to pylab, a matplotlib-based Python environment [backend: module://IPython.kernel.zmq.pylab.backend_inline].\n", "For more information, type 'help(pylab)'.\n" ] } ], "prompt_number": 1 }, { "cell_type": "code", "collapsed": false, "input": [ "from brian import *\n", "from brian.library.modelfitting import *\n", "from brian.library.random_processes import *" ], "language": "python", "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [ { "output_type": "stream", "stream": "stderr", "text": [ "D:\\SVN\\Brian\\brian\\utils\\sparse_patch\\__init__.py:39: UserWarning: Couldn't find matching sparse matrix patch for scipy version 0.12.0, but in most cases this shouldn't be a problem.\n", " warnings.warn(\"Couldn't find matching sparse matrix patch for scipy version %s, but in most cases this shouldn't be a problem.\" % scipy.__version__)\n" ] } ], "prompt_number": 2 }, { "cell_type": "code", "collapsed": false, "input": [ "R, tau = 1., 5 * ms\n", "eqs = Equations('''dV/dt=(R*I-V)/tau : 1''')\n", "eqs += OrnsteinUhlenbeck('I', \n", " mu=.5, sigma=.5, tau=5 * ms)" ], "language": "python", "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "prompt_number": 3 }, { "cell_type": "code", "collapsed": false, "input": [ "G = NeuronGroup(N=1, model=eqs, reset=0., threshold=1., refractory=5 * ms)\n", "Mspikes = SpikeMonitor(G)\n", "Vstate = StateMonitor(G, 'V', record=True)\n", "Istate = StateMonitor(G, 'I', record=True)\n", "net = Network(G, Mspikes, Vstate, Istate)\n", "reinit_default_clock()\n", "net.run(1 * second)" ], "language": "python", "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [], "prompt_number": 4 }, { "cell_type": "code", "collapsed": false, "input": [ "figure(figsize=(10,4))\n", "Vif, current, spikes = Vstate[0], Istate[0], Mspikes.spiketimes[0]\n", "plot(Vstate.times, Vif, '-k');\n", "vlines(spikes, 1, 3);" ], "language": "python", "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "metadata": {}, "output_type": "display_data", "png": "iVBORw0KGgoAAAANSUhEUgAAAl0AAAEACAYAAACJesOfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEQizAAAAAElFTkSuQmCC\n", "text": [ "