diff --git a/docs/_static/NG-RC-vs-Traditional-RC.png b/docs/_static/NG-RC-vs-Traditional-RC.png index 87b2f0749..1fde73d1f 100644 Binary files a/docs/_static/NG-RC-vs-Traditional-RC.png and b/docs/_static/NG-RC-vs-Traditional-RC.png differ diff --git a/docs/quickstart/analysis.ipynb b/docs/quickstart/analysis.ipynb index cd25b8b3a..a468286d3 100644 --- a/docs/quickstart/analysis.ipynb +++ b/docs/quickstart/analysis.ipynb @@ -225,7 +225,7 @@ } }, "source": [ - "In the ``Bifurcation1D`` analyzer, ``model`` refers to the modelto be analyzed (essentially the analyzer will access the derivative function in the model), ``target_vars`` denotes the target variables, ``target_pars`` denotes the changing parameters, and ``resolution`` determines the resolutioin of the analysis." + "In the ``Bifurcation1D`` analyzer, ``model`` refers to the model to be analyzed (essentially the analyzer will access the derivative function in the model), ``target_vars`` denotes the target variables, ``target_pars`` denotes the changing parameters, and ``resolution`` determines the resolutioin of the analysis." ] }, { diff --git a/docs/quickstart/installation.rst b/docs/quickstart/installation.rst index 4a83e100f..08f3f59b4 100644 --- a/docs/quickstart/installation.rst +++ b/docs/quickstart/installation.rst @@ -118,17 +118,12 @@ Alternatively, you can download the preferred release ".whl" file for jaxlib, an pip install jax -.. warning:: - For m1 macOS users, you should run your python environment on ``Apple`` silicon instead of ``intel`` - silicon since ``rosetta2`` cannot translate ``jaxlib``. One suggestion is uninstall miniconda3 and install - miniforge3 for managing your python environment. - Windows ^^^^^^^ For **Windows** users, JAX can be installed by the following methods: -- **Method 1**: There are several community supported Windows build for jax, please refer +- **Method 1**: There are several communities support JAX for Windows, please refer to the github link for more details: https://github.com/cloudhan/jax-windows-builder . Simply speaking, the provided binary releases of JAX for Windows are available at https://whls.blob.core.windows.net/unstable/index.html . diff --git a/docs/quickstart/simulation.ipynb b/docs/quickstart/simulation.ipynb index 2acc444ba..4dc055d7f 100644 --- a/docs/quickstart/simulation.ipynb +++ b/docs/quickstart/simulation.ipynb @@ -103,7 +103,7 @@ } }, "source": [ - "A E-I balanced network is composed of two neuron groups and the synaptic connections between them. Specifically, they include:\n", + "An E-I balanced network is composed of two neuron groups and the synaptic connections between them. Specifically, they include:\n", "1. a group of excitatory neurons, $\\mathrm{E}$,\n", "2. a group of inhibitory neurons, $\\mathrm{I}$,\n", "3. synaptic connections within the excitatory and inhibitory neuron groups, respectively, and \n", @@ -131,15 +131,7 @@ "name": "#%%\n" } }, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:absl:No GPU/TPU found, falling back to CPU. (Set TF_CPP_MIN_LOG_LEVEL=0 and rerun for more info.)\n" - ] - } - ], + "outputs": [], "source": [ "E = bp.neurons.LIF(3200, V_rest=-60., V_th=-50., V_reset=-60.,\n", " tau=20., tau_ref=5., method='exp_auto',\n", @@ -159,7 +151,7 @@ } }, "source": [ - "When defining the LIF neuron group, the parameters can be tuned according to users' need. The first parameter denotes the number of neurons. Here the ratio of excitatory and inhibitory neurons is set 4:1. ``V_rest`` denotes the resting potential, ``V_th`` denotes the firing threshold, ``V_reset`` denotes the reset value after firing, ``tau`` is the time constant, and ``tau_ref`` is the duration of the refractory period. ``method`` refers to the numerical integration method to be used in simulation. " + "When defining the LIF neuron group, the parameters can be tuned according to users' need. The first parameter denotes the number of neurons. Here the ratio of excitatory and inhibitory neurons is set to 4:1. ``V_rest`` denotes the resting potential, ``V_th`` denotes the firing threshold, ``V_reset`` denotes the reset value after firing, ``tau`` is the time constant, and ``tau_ref`` is the duration of the refractory period. ``method`` refers to the numerical integration method to be used in simulation." ] }, { @@ -186,20 +178,20 @@ "outputs": [], "source": [ "E2E = bp.synapses.Exponential(E, E, bp.conn.FixedProb(prob=0.02), g_max=0.6,\n", - " tau=5., method='exp_auto',\n", - " output=bp.synouts.COBA(E=0.))\n", + " tau=5., output=bp.synouts.COBA(E=0.),\n", + " method='exp_auto')\n", "\n", "E2I = bp.synapses.Exponential(E, I, bp.conn.FixedProb(prob=0.02), g_max=0.6,\n", - " tau=5., method='exp_auto',\n", - " output=bp.synouts.COBA(E=0.))\n", + " tau=5., output=bp.synouts.COBA(E=0.),\n", + " method='exp_auto')\n", "\n", "I2E = bp.synapses.Exponential(I, E, bp.conn.FixedProb(prob=0.02), g_max=6.7,\n", - " tau=10., method='exp_auto',\n", - " output=bp.synouts.COBA(E=-80.))\n", + " tau=10., output=bp.synouts.COBA(E=-80.),\n", + " method='exp_auto')\n", "\n", "I2I = bp.synapses.Exponential(I, I, bp.conn.FixedProb(prob=0.02), g_max=6.7,\n", - " tau=10., method='exp_auto',\n", - " output=bp.synouts.COBA(E=-80.))" + " tau=10., output=bp.synouts.COBA(E=-80.),\n", + " method='exp_auto')" ] }, { @@ -211,7 +203,7 @@ } }, "source": [ - "Here we use the Exponential synapse model (``Exponential``) to simulate synaptic connections. Among the parameters of the model, the first two denotes the pre- and post-synaptic neuron groups, respectively. The third one refers to the connection types. In this example, we use ``bp.conn.FixedProb``, which connects the presynaptic neurons to postsynaptic neurons with a given probability (detailed information is available in [Synaptic Connection](../tutorial_toolbox/synaptic_connections.ipynb)). The following three parameters describes the dynamic properties of the synapse, and the last one is the numerical integration method as that in the LIF model." + "Here we use the Exponential synapse model (``bp.synapses.Exponential``) to simulate synaptic connections. Among the parameters of the model, the first two denotes the pre- and post-synaptic neuron groups, respectively. The third one refers to the connection types. In this example, we use ``bp.conn.FixedProb``, which connects the presynaptic neurons to postsynaptic neurons with a given probability (detailed information is available in [Synaptic Connection](../tutorial_toolbox/synaptic_connections.ipynb)). The following three parameters describes the dynamic properties of the synapse, and the last one is the numerical integration method as that in the LIF model." ] }, { @@ -275,7 +267,7 @@ } }, "source": [ - "After build a SNN, we can use it for dynamic simulation. To run a simulation, we need first wrap the network model into a **runner**. Currently BrainPy provides ``DSRunner`` and ``ReportRunner`` in ``brainpy.dyn``, which will be expanded in the [Runners](../tutorial_simulation/runner.ipynb) tutorial. Here we use ``DSRunner`` as an example:" + "After building a SNN, we can use it for dynamic simulation. To run a simulation, we need to wrap the network model into a **runner** first. BrainPy provides ``DSRunner`` in ``brainpy.dyn``, which will be expanded in the [Runners](../tutorial_simulation/runner.ipynb) tutorial. Users can initialize ``DSRunner`` as followed:" ] }, { @@ -304,7 +296,7 @@ } }, "source": [ - "To make dynamic simulation more applicable and powerful, users can [**monitor**](../tutorial_toolbox/monitors.ipynb) variable trajectories and give [**inputs**](../tutorial_toolbox/inputs.ipynb) to target neuron groups. Here we monitor the ``spike`` variable in the ``E`` and ``I`` LIF model, which refers to the spking status of the neuron group, and give a constant input to both neuron groups. The time interval of numerical integration ``dt`` (with the default value of 0.1) can also be specified.\n", + "To make dynamic simulation more applicable and powerful, users can monitor variable trajectories and give inputs to target neuron groups. Here we monitor the ``spike`` variable in the ``E`` and ``I`` LIF model, which refers to the spking status of the neuron group, and give a constant input to both neuron groups. The time interval of numerical integration ``dt`` (with the default value of 0.1) can also be specified.\n", "\n", "More details of how to give inputs and monitors please refer to [Dynamic Simulation](../tutorial_simulation/index.rst).\n", "\n", @@ -327,7 +319,7 @@ "application/vnd.jupyter.widget-view+json": { "version_major": 2, "version_minor": 0, - "model_id": "98a9eee1680745d8bf9999d16732067e" + "model_id": "e728df4a34cf46a4914db2455f0fadcc" } }, "metadata": {}, @@ -354,7 +346,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 8, "id": "f3aab08c", "metadata": { "pycharm": { @@ -365,7 +357,7 @@ { "data": { "text/plain": "
", - "image/png": "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\n" + "image/png": "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\n" }, "metadata": { "needs_background": "light" @@ -379,6 +371,7 @@ "plt.figure(figsize=(12, 4.5))\n", "\n", "plt.subplot(121)\n", + "\n", "bp.visualize.raster_plot(runner.mon.ts, runner.mon['E.spike'], show=False)\n", "plt.subplot(122)\n", "bp.visualize.raster_plot(runner.mon.ts, runner.mon['I.spike'], show=True)" @@ -448,7 +441,7 @@ "source": [ "To construct a decision-making network, we should build all neuron groups:\n", "\n", - "1. Two excicatory neuron groups with different selectivity, $\\mathrm{A}$ and $\\mathrm{B}$, and other excicatory neurons, $\\mathrm{N}$;\n", + "1. Two excitatory neuron groups with different selectivity, $\\mathrm{A}$ and $\\mathrm{B}$, and other excitatory neurons, $\\mathrm{N}$;\n", "2. An inhibitory neuron group, $\\mathrm{I}$;\n", "3. Neurons generating external inputs $\\mathrm{I_A}$ and $\\mathrm{I_B}$;\n", "4. Neurons generating noise to all neuron groups, $\\mathrm{noise_A}$, $\\mathrm{noise_B}$, $\\mathrm{noise_N}$, and $\\mathrm{noise_I}$.\n", @@ -458,7 +451,7 @@ "1. Connection from excitatory neurons to others, $\\mathrm{A2A}$, $\\mathrm{A2B}$, $\\mathrm{A2N}$, $\\mathrm{A2I}$, $\\mathrm{B2A}$, $\\mathrm{B2B}$, $\\mathrm{B2N}$, $\\mathrm{B2I}$, and $\\mathrm{N2A}$, $\\mathrm{N2B}$, $\\mathrm{N2N}$, $\\mathrm{N2I}$;\n", "2. Connection from inhibitory neurons to others, $\\mathrm{I2A}$, $\\mathrm{I2B}$, $\\mathrm{I2N}$, $\\mathrm{I2I}$;\n", "3. Connection from external inputs to selective neuron groups, $\\mathrm{IA2A}$, $\\mathrm{IB2B}$;\n", - "4. Connectioni from noise neurons to excitatory and inhibitory neurons, $\\mathrm{noise2A}$, $\\mathrm{noise2B}$, $\\mathrm{noise2N}$, $\\mathrm{noise2I}$." + "4. Connection from noise neurons to excitatory and inhibitory neurons, $\\mathrm{noise2A}$, $\\mathrm{noise2B}$, $\\mathrm{noise2N}$, $\\mathrm{noise2I}$." ] }, { @@ -501,7 +494,7 @@ } }, "source": [ - "To build $\\mathrm{I_A}$ and $\\mathrm{I_B}$, we shall define a class of neuron groups that can generate stochastic Possion stimulu. Two define neuron groups, they should inherit `brainpy.dyn.NeuGroup`." + "To build $\\mathrm{I_A}$ and $\\mathrm{I_B}$, we shall define a class of neuron groups that can generate stochastic Possion stimulu. To define neuron groups, they should inherit from `brainpy.dyn.NeuGroup`." ] }, { @@ -1224,7 +1217,7 @@ { "cell_type": "markdown", "source": [ - "In genral, a dataset for whole-brain modeling consists of the following parts:\n", + "In general, a dataset for whole-brain modeling consists of the following parts:\n", "\n", "1\\. A structural connectivity matrix which captures the synaptic connection strengths between brain areas. It often derived from DTI tractography of the whole brain. The connectome is then typically parcellated in a preferred atlas (for example the AAL2 atlas) and the number of axonal fibers connecting each brain area with every other area is counted. This number serves as an indication of the synaptic coupling strengths between the areas of the brain.\n", "\n", diff --git a/docs/quickstart/training.ipynb b/docs/quickstart/training.ipynb index 7d5228dbb..892db04f1 100644 --- a/docs/quickstart/training.ipynb +++ b/docs/quickstart/training.ipynb @@ -33,7 +33,7 @@ } }, "source": [ - "In recent years, we saw the revolution that training a dynamical system from data or tasks has provided important insights to understand brain functions. To support this, BrainPy porvides various interfaces to help users train dynamical systems. " + "In recent years, we saw the revolution that training a dynamical system from data or tasks has provided important insights to understand brain functions. To support this, BrainPy provides various interfaces to help users train dynamical systems." ] }, { @@ -743,7 +743,7 @@ } }, "source": [ - "Here, we demonstrate how to train an artificial recurrent neural network by using a white noise integration task. In this task, we want our trained RNN model has the ability to integrate white noise. For example, if we has a time series of noise data, " + "Here, we demonstrate how to train an artificial recurrent neural network by using a white noise integration task. In this task, we want our trained RNN model has the ability to integrate white noise. For example, if we have a time series of noise data," ] }, { diff --git a/docs/tutorial_analysis/decision_making_model.ipynb b/docs/tutorial_analysis/decision_making_model.ipynb index 37e8c754e..8d0ddc7d0 100644 --- a/docs/tutorial_analysis/decision_making_model.ipynb +++ b/docs/tutorial_analysis/decision_making_model.ipynb @@ -33,7 +33,7 @@ } }, "source": [ - "In this section, we are going to use the [low-dimensional analyzers](./lowdim_analysis.ipynb) to make phase plane and bifurcation analysis for the decision making model proposed by (Wong & Wang) [1]. " + "In this section, we are going to use the [low-dimensional analyzers](./lowdim_analysis.ipynb) to make phase plane and bifurcation analysis for the decision-making model proposed by (Wong & Wang) [1]." ] }, { @@ -203,7 +203,7 @@ } }, "source": [ - "The advantage of the reduced model is that we can understand what dynamical behaviors the model generate for a particular parmeter set using phase-plane analysis and the explore how this behavior changed when the model parameters are variaed (bifurcation analysis). \n", + "The advantage of the reduced model is that we can understand what dynamical behaviors the model generate for a particular parmeter set using phase-plane analysis and the explore how this behavior changed when the model parameters are varied (bifurcation analysis).\n", "\n", "To this end, we will use ``brainpy.analysis`` module. " ] @@ -517,7 +517,7 @@ } }, "source": [ - "To see how the ohase portrait of the system changed when we chang the stimulus current, we will generate a bifucation diagram for the reduced model. On the bifurcation diagram the fixed points of the model are shown as a function of a changing parameter. \n", + "To see how the phase portrait of the system changed when we chang the stimulus current, we will generate a bifurcation diagram for the reduced model. On the bifurcation diagram the fixed points of the model are shown as a function of a changing parameter.\n", "\n", "In the next, we generate bifurcation diagrams with the different parameters. " ] diff --git a/docs/tutorial_analysis/highdim_analysis.ipynb b/docs/tutorial_analysis/highdim_analysis.ipynb index 74b69cc64..bd34ecbe6 100644 --- a/docs/tutorial_analysis/highdim_analysis.ipynb +++ b/docs/tutorial_analysis/highdim_analysis.ipynb @@ -206,7 +206,7 @@ } }, "source": [ - "``brainpy.analysis.SlowPointFinder`` is aimed to find slow/fixed points of high-dimensional systems. Of course, it can optimize to find fixed points of low-dimensional systems. We take the 2D decision making system as an example. " + "``brainpy.analysis.SlowPointFinder`` is aimed to find slow/fixed points of high-dimensional systems. Of course, it can optimize to find fixed points of low-dimensional systems. We take the 2D decision-making system as an example." ] }, { @@ -515,7 +515,7 @@ { "cell_type": "markdown", "source": [ - "The following code demonstrates how to use ``SlowPointFinder`` to find fixed points of a continous-attractor neural network." + "The following code demonstrates how to use ``SlowPointFinder`` to find fixed points of a continuous attractor neural network." ], "metadata": { "collapsed": false, @@ -569,7 +569,7 @@ { "cell_type": "markdown", "source": [ - "The found fixed points are a series of attractor. We can visualized this line of attractors on a 2D space." + "The found fixed points are a series of attractor. We can visualize this line of attractors on a 2D space." ], "metadata": { "collapsed": false, diff --git a/docs/tutorial_analysis/lowdim_analysis.ipynb b/docs/tutorial_analysis/lowdim_analysis.ipynb index ad86ecd60..ebe4dffff 100644 --- a/docs/tutorial_analysis/lowdim_analysis.ipynb +++ b/docs/tutorial_analysis/lowdim_analysis.ipynb @@ -121,7 +121,7 @@ } }, "source": [ - "As known to us all, this functuon has multiple fixed points ($\\frac{dx}{dt} = 0$) when $I=0$." + "As known to us all, this function has multiple fixed points ($\\frac{dx}{dt} = 0$) when $I=0$." ] }, { @@ -269,7 +269,7 @@ } }, "source": [ - "Yeah, absolutelty, ``brainpy.analysis.PhasePlane1D`` gives us the right fixed points, and correctly evalutes the stability of these fixed points. " + "Yeah, absolutelty, ``brainpy.analysis.PhasePlane1D`` gives us the right fixed points, and correctly evaluates the stability of these fixed points." ] }, { @@ -280,7 +280,7 @@ } }, "source": [ - "Phase plane is important, because it give us the intuitive understanding how the system evolves with the given parameters. However, in most cases where we care about how the parameters affect the system behaviors, we should make bifurcation analysis. [brainpy.analysis.Bifurcation1D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.Bifurcation1D.rst) is a convenient interface to help you get the insights of how the dynamics of a 1D system changes with parameters. \n", + "Phase plane is important, because it gives us the intuitive understanding how the system evolves with the given parameters. However, in most cases where we care about how the parameters affect the system behaviors, we should make bifurcation analysis. [brainpy.analysis.Bifurcation1D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.Bifurcation1D.rst) is a convenient interface to help you get the insights of how the dynamics of a 1D system changes with parameters.\n", "\n", "Similar to ``brainpy.analysis.PhasePlane1D``, ``brainpy.analysis.Bifurcation1D`` receives arguments like \"model\", \"target_vars\", \"pars_update\", and \"resolutions\". Besides, one more important argument **\"target_pars\"** should be provided, which specifies the range of the target parameter in bifurcation analysis." ] @@ -293,7 +293,7 @@ } }, "source": [ - "Here, we systematically change the parameter \"Iext\" from 0 to 1.5. According to the bifurcation theory, we know this simple system has a fold bifurcation when $I=1.0$. Because at $I=1.0$, two fixed points collide with each other into a saddle point and then disappear. Does BrainPy's analysis toolkit ``brainpy.analysis.Bifurcation1D`` is capable of performing these analysis? Let's make a try." + "Here, we systematically change the parameter \"Iext\" from 0 to 1.5. According to the bifurcation theory, we know this simple system has a fold bifurcation when $I=1.0$. Because at $I=1.0$, two fixed points collide with each other into a saddle point and then disappear. Does BrainPy's analysis toolkit ``brainpy.analysis.Bifurcation1D`` is capable of performing these analyses? Let's make a try." ] }, { @@ -355,7 +355,7 @@ } }, "source": [ - "It is worthy to note that bifurcation analysis in BrainPy is hard to find out the saddle point (when $I=0$ for this system). This is because the saddle point at the bifurcation just exists at a moment. While the numerical method used in BrainPy analysis toolkit is almost impossible to evaluate the point exactly at the saddle. However, if the user has the minimal knowledge about the bifurcation theory, saddle point (the collision point of two fixed points) can be easily infered from the fixed point evolution." + "It is worthy to note that bifurcation analysis in BrainPy is hard to find out the saddle point (when $I=0$ for this system). This is because the saddle point at the bifurcation just exists at a moment. While the numerical method used in BrainPy analysis toolkit is almost impossible to evaluate the point exactly at the saddle. However, if the user has the minimal knowledge about the bifurcation theory, saddle point (the collision point of two fixed points) can be easily inferred from the fixed point evolution." ] }, { @@ -388,10 +388,10 @@ } }, "source": [ - "Phase plane analysis is one of the most important techniques for studying the behavior of nonlinear systems, since there is usually no analytical solution for a nonlinear system. BrainPy can help users to plot phase plane of 1D systems or 2D systems. Specifically, we provides [brainpy.analysis.PhasePlane1D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.PhasePlane1D.rst) and [brainpy.analysis.PhasePlane2D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.PhasePlane2D.rst). It can help to plot:\n", + "Phase plane analysis is one of the most important techniques for studying the behavior of nonlinear systems, since there is usually no analytical solution for a nonlinear system. BrainPy can help users to plot phase plane of 1D systems or 2D systems. Specifically, we provide [brainpy.analysis.PhasePlane1D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.PhasePlane1D.rst) and [brainpy.analysis.PhasePlane2D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.PhasePlane2D.rst). It can help to plot:\n", "\n", "- **Nullcline**: The zero-growth isoclines, such as $g(x, y)=0$ and $g(x, y)=0$.\n", - "- **Fixed points**: The equilibrium points of the system, which are located at all of the nullclines intersect.\n", + "- **Fixed points**: The equilibrium points of the system, which are located at all the nullclines intersect.\n", "- **Vector field**: The vector field of the system.\n", "- **Limit cycles**: The limit cycles.\n", "- **Trajectories**: A simulation trajectory with the given initial values." @@ -710,7 +710,7 @@ "It is highly recommended that users specify the resolution to specific parameters or variables by a dict rather than set a float value, which will be applied to all variables. Otherwise, the computation will occupy too much memory if the resolution is set very small. For example, if you want to set the resolution of variable `x` as 0.01, please use `resolutions={'x': 0.01}`.\n", "```\n", "\n", - "Enabling set ``resolutions`` with a tensor will give the user the maximal flexibility. Usually, the numerical alalysis does not work well at inflection points. Therefore, we can increase the granularity near the inflection points. For example, if there is an inflextion point at $1$, we can set the resolution with:" + "Enabling set ``resolutions`` with a tensor will give the user the maximal flexibility. Usually, the numerical analysis does not work well at inflection points. Therefore, we can increase the granularity near the inflection points. For example, if there is an inflection point at $1$, we can set the resolution with:" ] }, { @@ -825,7 +825,7 @@ } }, "source": [ - "We will first see the codimension 1 bifurcation anlysis of the model. For example, we vary the input $I_{ext}$ between 0 to 1 and see how the system change it's stability." + "We will first see the codimension 1 bifurcation analysis of the model. For example, we vary the input $I_{ext}$ between 0 and 1 and see how the system change its stability." ] }, { @@ -996,7 +996,7 @@ "BrainPy also provides a tool for fast-slow system bifurcation analysis by using [brainpy.analysis.FastSlow1D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.FastSlow1D.rst) and [brainpy.analysis.FastSlow2D](../apis/auto/analysis/generated/brainpy.analysis.lowdim.FastSlow2D.rst). This method is proposed by John Rinzel [1, 2, 3]. (J Rinzel, 1985, 1986, 1987) proposed that in a fast-slow dynamical system, we can treat the slow variables as the bifurcation parameters, and then study how the different value of slow variables affect the bifurcation of the fast sub-system.\n", "\n", "\n", - "Fast-slow bifurcation methods are very usefull in the bursting neuron analysis. I will illustrate this by using the Hindmarsh-Rose model. The Hindmarsh–Rose model of neuronal activity is aimed to study the spiking-bursting behavior of the membrane potential observed in experiments made with a single neuron. Its dynamics are governed by:\n", + "Fast-slow bifurcation methods are very useful in the bursting neuron analysis. I will illustrate this by using the Hindmarsh-Rose model. The Hindmarsh–Rose model of neuronal activity is aimed to study the spiking-bursting behavior of the membrane potential observed in experiments made with a single neuron. Its dynamics are governed by:\n", "\n", "$$\n", "\\begin{aligned}\n", @@ -1015,7 +1015,7 @@ } }, "source": [ - "First of all, let's define the Hindmarsh–Rose model with BrainPy. " + "First, let's define the Hindmarsh–Rose model with BrainPy." ] }, { diff --git a/docs/tutorial_building/build_network_models.ipynb b/docs/tutorial_building/build_network_models.ipynb index e031fcf64..66545a958 100644 --- a/docs/tutorial_building/build_network_models.ipynb +++ b/docs/tutorial_building/build_network_models.ipynb @@ -87,7 +87,7 @@ } }, "source": [ - "The E-I balanced network was first proposed to explain the irregular firing patterns of cortical neurons and comfirmed by experimental data. The network [1] we are going to implement consists of excitatory (E) neurons and inhibitory (I) neurons, the ratio of which is about 4 : 1. The biggest difference between excitatory and inhibitory neurons is the reversal potential - the reversal potential of inhibitory neurons is much lower than that of excitatory neurons. Besides, the membrane time constant of inhibitory neurons is longer than that of excitatory neurons, which indicates that inhibitory neurons have slower dynamics." + "The E-I balanced network was first proposed to explain the irregular firing patterns of cortical neurons and confirmed by experimental data. The network [1] we are going to implement consists of excitatory (E) neurons and inhibitory (I) neurons, the ratio of which is about 4 : 1. The biggest difference between excitatory and inhibitory neurons is the reversal potential - the reversal potential of inhibitory neurons is much lower than that of excitatory neurons. Besides, the membrane time constant of inhibitory neurons is longer than that of excitatory neurons, which indicates that inhibitory neurons have slower dynamics." ] }, { @@ -395,7 +395,7 @@ } }, "source": [ - "All elements are passed as ``**kwargs`` argument can be accessed by the provided keys. This will affect the following dynamics simualtion and will be discussed in greater detail in tutorial of [Runners](../tutorial_toolbox/runners.ipynb)." + "All elements are passed as ``**kwargs`` argument can be accessed by the provided keys. This will affect the following dynamics simulation and will be discussed in greater detail in tutorial of [Runners](../tutorial_toolbox/runners.ipynb)." ] }, { diff --git a/docs/tutorial_building/build_synapse_models.ipynb b/docs/tutorial_building/build_synapse_models.ipynb index c83628c2c..f39967e84 100644 --- a/docs/tutorial_building/build_synapse_models.ipynb +++ b/docs/tutorial_building/build_synapse_models.ipynb @@ -27,7 +27,7 @@ { "cell_type": "markdown", "source": [ - "In BrainPy, synapse models can be created by several ways. In this section, we will talk about a structural building process with ``brainpy.dyn.TwoEndConn``. ``brainpy.dyn.TwoEndConn`` is used to create models with pre- and post-synaptic neuron groups.\n", + "In BrainPy, synapse models can be created by several ways. In this section, we will talk about a structural building process with ``brainpy.dyn.TwoEndConn``, which is used to create models with pre- and post-synaptic neuron groups.\n", "A synapse model is decomposed into several components in ``brainpy.dyn.TwoEndConn``. In such a way, building a synapse model can follow a modular and composable programming interface.\n", "Fore more details of defining a general synapse model, please refer to ``brainpy.dyn.SynConn`` in [Tutorial: Customizing Synapse Models](./customize_synapse_models.ipynb)." ], @@ -132,7 +132,7 @@ "I_{\\mathrm{post}}(t) = g_{\\mathrm{max}} * g * (V_{\\mathrm{post}}(t)-E),\n", "$$\n", "\n", - "where its $f_{\\mathrm{dyn}}$ is defined as equation (1), its $f_{\\mathrm{LTP}}$ and $f_{\\mathrm{STP}}$ is the identy function $x = f(x)$, $f_{\\mathrm{out}}$ is defined as a conductance-based form with $(V(t)-E_{\\mathrm{post}})$.\n", + "where its $f_{\\mathrm{dyn}}$ is defined as equation (1), its $f_{\\mathrm{LTP}}$ and $f_{\\mathrm{STP}}$ is the identity function $x = f(x)$, $f_{\\mathrm{out}}$ is defined as a conductance-based form with $(V_{\\mathrm{post}}(t)-E)$.\n", "\n" ], "metadata": { diff --git a/docs/tutorial_building/customize_dynamical_systems.ipynb b/docs/tutorial_building/customize_dynamical_systems.ipynb index 6ebe397e8..ba0d8772c 100644 --- a/docs/tutorial_building/customize_dynamical_systems.ipynb +++ b/docs/tutorial_building/customize_dynamical_systems.ipynb @@ -32,7 +32,7 @@ "source": [ "The previous sections have shown how to build neuron models, synapse models, and network models. In fact, these brain objects all inherit the base class [brainpy.dyn.DynamicalSystem](../apis/auto/dyn/generated/brainpy.dyn.base.DynamicalSystem.rst), ``brainpy.dyn.DynamicalSystem`` is the universal language to define dynamical models in BrainPy.\n", "\n", - "To begin with, let's make a rief summary of previous dynamic models and give the definition of a dynamical system." + "To begin with, let's make a brief summary of previous dynamic models and give the definition of a dynamical system." ] }, { diff --git a/docs/tutorial_building/customize_neuron_models.ipynb b/docs/tutorial_building/customize_neuron_models.ipynb index 53f6c1aeb..28a3b9d77 100644 --- a/docs/tutorial_building/customize_neuron_models.ipynb +++ b/docs/tutorial_building/customize_neuron_models.ipynb @@ -78,7 +78,7 @@ "Discontinuous evolution may be triggered by events, such as the reset of membrane potential. \n", "Moreover, it is common in a neural system that a dynamical system has different states, such as the excitable or refractory\n", "state in a [leaky integrate-and-fire (LIF) model](https://brainmodels.readthedocs.io/en/latest/apis/generated/brainmodels.neurons.LIF.html). \n", - "In this section, we will use two examples to illustrate how to capture these complexity in neuron modeling." + "In this section, we will use two examples to illustrate how to capture these complexities in neuron modeling." ] }, { @@ -130,7 +130,7 @@ } }, "source": [ - "The Hodgkin-Huxley (HH) model is a continuous-time dynamical system. It is one of the most successful mathematical models of a complex biological process that has ever been formulated. Changes of the membrane potential influence the conductances of different channels, elaborately modeling the neural activities in biological systems. Mathematically, the model is given by:\n", + "The Hodgkin-Huxley (HH) model is a continuous-time dynamical system. It is one of the most successful mathematical models of a complex biological process that has ever been formulated. Changes of the membrane potential influence the conductance of different channels, elaborately modeling the neural activities in biological systems. Mathematically, the model is given by:\n", "\n", "$$\n", "\\begin{aligned}\n", @@ -146,7 +146,7 @@ "\\end{aligned}\n", "$$\n", "\n", - "where $V$ is the membrane potential, $C_m$ is the membrane capacitance per unit area, $E_K$ and $E_{Na}$ are the potassium and sodium reversal potentials, respectively, $E_l$ is the leak reversal potential, $\\bar{g}_K$ and $\\bar{g}_{Na}$ are the potassium and sodium conductances per unit area, respectively, and $\\bar{g}_l$ is the leak conductance per unit area. Because the potassium and sodium channels are voltage-sensitive, according to the biological experiments, $m$, $n$ and $h$ are used to simulate the activation of the channels. Speficially, $n$ measures the activatio of potassium channels, and $m$ and $h$ measures the activation and inactivation of sodium channels, respectively. $\\alpha_{x}$ and $\\beta_{x}$ are rate constants for the ion channel x and depend exclusively on the membrane potential." + "where $V$ is the membrane potential, $C_m$ is the membrane capacitance per unit area, $E_K$ and $E_{Na}$ are the potassium and sodium reversal potentials, respectively, $E_l$ is the leak reversal potential, $\\bar{g}_K$ and $\\bar{g}_{Na}$ are the potassium and sodium conductance per unit area, respectively, and $\\bar{g}_l$ is the leak conductance per unit area. Because the potassium and sodium channels are voltage-sensitive, according to the biological experiments, $m$, $n$ and $h$ are used to simulate the activation of the channels. Specially, $n$ measures the activation of potassium channels, and $m$ and $h$ measures the activation and inactivation of sodium channels, respectively. $\\alpha_{x}$ and $\\beta_{x}$ are rate constants for the ion channel x and depend exclusively on the membrane potential." ] }, { diff --git a/docs/tutorial_building/customize_synapse_models.ipynb b/docs/tutorial_building/customize_synapse_models.ipynb index b8896d26a..10bcf390f 100644 --- a/docs/tutorial_building/customize_synapse_models.ipynb +++ b/docs/tutorial_building/customize_synapse_models.ipynb @@ -33,7 +33,7 @@ } }, "source": [ - "Synaptic computation is the core of brain dynamics programming. This is beacuse in a real project most of the simulation time spends on the computation of synapses. In order to achieve efficient synaptic computation, BrainPy provides many useful supports. Here, we are going to explore the details of these supports. " + "Synaptic computation is the core of brain dynamics programming. This is because in a real project most of the simulation time spends on the computation of synapses. In order to achieve efficient synaptic computation, BrainPy provides many useful supports. Here, we are going to explore the details of these supports." ] }, { @@ -263,7 +263,7 @@ } }, "source": [ - "As seen in the above two synapse models, synaptic computations are usually involved with variable delays. A delay time (typically 0.3–0.5 ms) is usually required for a neurotransmitter to be released from a presynaptic membrane, diffuse across the synaptic cleft, and bind to a receptor site on the post-synaptic membrane.\n", + "As seen in the above two synapse models, synaptic computations are usually involved with variable delays. A delay time (typically 0.3–0.5 ms) is usually required for a neurotransmitter to be released from a pre-synaptic membrane, diffuse across the synaptic cleft, and bind to a receptor site on the post-synaptic membrane.\n", "\n", "BrainPy provides several kinds of delay variables for users, including:\n", "\n", @@ -451,7 +451,7 @@ } }, "source": [ - "Synaptic computations usually need to create connection between groups. BrainPy provides many wonderful supports to construct [synaptic connections](./synaptic_connections.ipynb). Simply speaking, ``brainpy.conn.Connector`` can create various data sturctures you want through the ``require()`` function. Take the random connection ``brainpy.conn.FixedProb`` which will be used in follows as the example, " + "Synaptic computations usually need to create connection between groups. BrainPy provides many wonderful supports to construct [synaptic connections](./synaptic_connections.ipynb). Simply speaking, ``brainpy.conn.Connector`` can create various data structures you want through the ``require()`` function. Take the random connection ``brainpy.conn.FixedProb`` which will be used in follows as the example," ] }, { @@ -632,7 +632,7 @@ } }, "source": [ - "Before we start, we need to define some useful helper functions to define and show synapse models. Then, we will highlight the key differences of model difinition when using different synaptic connections. " + "Before we start, we need to define some useful helper functions to define and show synapse models. Then, we will highlight the key differences of model definition when using different synaptic connections." ] }, { @@ -1068,7 +1068,7 @@ } }, "source": [ - "Sometimes, we can define some synapse models with special connection types, such as all-to-all connection, or one-to-one connection. For these special situations, even the connection information can be ignored, i.e., we do not need ``conn_mat`` or other structures any more. " + "Sometimes, we can define some synapse models with special connection types, such as all-to-all connection, or one-to-one connection. For these special situations, even the connection information can be ignored, i.e., we do not need ``conn_mat`` or other structures anymore." ] }, { @@ -1241,7 +1241,7 @@ } }, "source": [ - "Actually, the synaptic computation with these special connections can be very efficient! A concrete example please see a [decision making spiking model](https://brainpy-examples.readthedocs.io/en/latest/decision_making/Wang_2002_decision_making_spiking.html) in BrainPy-Examples. This implementation achievew at least four times acceleration comparing to the implementation in other frameworks. " + "Actually, the synaptic computation with these special connections can be very efficient! A concrete example please see a [decision making spiking model](https://brainpy-examples.readthedocs.io/en/latest/decision_making/Wang_2002_decision_making_spiking.html) in BrainPy-Examples. This implementation achieve at least four times acceleration comparing to the implementation in other frameworks." ] }, { diff --git a/docs/tutorial_building/overview_of_dynamic_model.ipynb b/docs/tutorial_building/overview_of_dynamic_model.ipynb index 5375fb15f..4ca7640cd 100644 --- a/docs/tutorial_building/overview_of_dynamic_model.ipynb +++ b/docs/tutorial_building/overview_of_dynamic_model.ipynb @@ -533,7 +533,7 @@ { "cell_type": "markdown", "source": [ - "However, in BrainPy, the built-in synapse models only support homogenous synaptic parameters, like the time constant $\\tau$. Users can [customize their synaptic models](./synapse_models.ipynb) when they want heterogeneous synatic parameters." + "However, in BrainPy, the built-in synapse models only support homogenous synaptic parameters, like the time constant $\\tau$. Users can [customize their synaptic models](./synapse_models.ipynb) when they want heterogeneous synaptic parameters." ], "metadata": { "collapsed": false, @@ -569,7 +569,7 @@ { "cell_type": "markdown", "source": [ - "In BrainPy, all the dynamically changed variables (no matter it is changed inside or outside of a jitted function) should be marked as ``brainpy.math.Variable``. BrainPy's built-in models also support modifying model parameters during simulation." + "In BrainPy, all the dynamically changed variables (no matter it is changed inside or outside the jitted function) should be marked as ``brainpy.math.Variable``. BrainPy's built-in models also support modifying model parameters during simulation." ], "metadata": { "collapsed": false, diff --git a/docs/tutorial_simulation/parallel_computing.ipynb b/docs/tutorial_simulation/parallel_computing.ipynb index 98e869f7c..8a69e12d9 100644 --- a/docs/tutorial_simulation/parallel_computing.ipynb +++ b/docs/tutorial_simulation/parallel_computing.ipynb @@ -43,12 +43,19 @@ "## Thread-based parallelization" ] }, + { + "cell_type": "markdown", + "source": [ + "### Multi-threading supported by `joblib`" + ], + "metadata": { + "collapsed": false + } + }, { "cell_type": "markdown", "metadata": {}, "source": [ - "Multi-threaded running of BrainPy models can be easily achieved.\n", - "\n", "The first approach is directly using the Python's multi-threading support. The following pseudocode demonstrates that by utilizing the `threading` backend of `joblib` library, we can easily achieve parallel execution based on multiple Python threads. However, the multi-threading parallelization based on this approach will get stuck in the well-known issue of Global Interpreter Lock (GIL) of Python." ] }, @@ -174,6 +181,15 @@ "plt.show()" ] }, + { + "cell_type": "markdown", + "source": [ + "### Multi-threading supported by `jax.vmap`" + ], + "metadata": { + "collapsed": false + } + }, { "cell_type": "markdown", "source": [ diff --git a/docs/tutorial_simulation/simulation_dsrunner.ipynb b/docs/tutorial_simulation/simulation_dsrunner.ipynb index 53dbf45df..998325034 100644 --- a/docs/tutorial_simulation/simulation_dsrunner.ipynb +++ b/docs/tutorial_simulation/simulation_dsrunner.ipynb @@ -256,7 +256,16 @@ { "cell_type": "markdown", "source": [ - "In BrainPy, any instance of ``brainpy.dyn.DSRunner`` has a built-in monitor. Users can set up a monitor when initializing a runner. There are multiple methods to initialize a monitor. The first method is to initialize a monitor is through a list of strings:" + "In BrainPy, any instance of ``brainpy.dyn.DSRunner`` has a built-in monitor. Users can set up a monitor when initializing a runner. There are multiple methods to initialize a monitor. The first method is to initialize a monitor is through a list of strings." + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "markdown", + "source": [ + "### Initialization with a list of strings" ], "metadata": { "collapsed": false @@ -359,6 +368,15 @@ } } }, + { + "cell_type": "markdown", + "source": [ + "### Initialization with index specification" + ], + "metadata": { + "collapsed": false + } + }, { "cell_type": "markdown", "source": [ @@ -411,6 +429,15 @@ } } }, + { + "cell_type": "markdown", + "source": [ + "### Explicit monitor target" + ], + "metadata": { + "collapsed": false + } + }, { "cell_type": "markdown", "source": [ @@ -465,7 +492,16 @@ { "cell_type": "markdown", "source": [ - "The forth method is similar to the third one, with the difference that the index specification is added:" + "### Explicit monitor target with index specification" + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "markdown", + "source": [ + "The fourth method is similar to the third one, with the difference that the index specification is added:" ], "metadata": { "collapsed": false @@ -620,7 +656,16 @@ { "cell_type": "markdown", "source": [ - "The first example is providing static input. The excitation and inhibition neurons all receive the same current intensity:" + "### Static inputs" + ], + "metadata": { + "collapsed": false + } + }, + { + "cell_type": "markdown", + "source": [ + "The first example is providing static inputs. The excitation and inhibition neurons all receive the same current intensity:" ], "metadata": { "collapsed": false @@ -668,6 +713,15 @@ } } }, + { + "cell_type": "markdown", + "source": [ + "### Iterable inputs" + ], + "metadata": { + "collapsed": false + } + }, { "cell_type": "markdown", "source": [