Merge pull request #177 from PKU-NIP-Lab/updates

Updates and fixes

.gitignore

@@ -5,6 +5,7 @@ publishment.md
 
 development
 
+examples/simulation/data
 examples/analysis/data
 extensions/.idea
 extensions/wheelhouse

README.md

@@ -27,7 +27,7 @@ BrainPy is a flexible, efficient, and extensible framework for computational neu
 
 
 
-## Installation
+## Install
 
 BrainPy is based on Python (>=3.7) and can be installed on  Linux (Ubuntu 16.04 or later), macOS (10.12 or later), and Windows platforms. Install the latest version of BrainPy:
 
@@ -42,239 +42,6 @@ The following packages are required for ``BrainPy``:
 For detailed installation instructions, please refer to the documentation: [Quickstart/Installation](https://brainpy.readthedocs.io/en/latest/quickstart/installation.html)
 
 
-
-## Examples
-
-
-
-```python
-import brainpy as bp
-```
-
-
-
-### 1. Operator level
-
-Mathematical operators in BrainPy are the same as those in NumPy. 
-
-```python
->>> import numpy as np
->>> import brainpy.math as bm
-
-# array creation
->>> np_arr = np.zeros((2, 4));  np_arr  
-array([[0., 0., 0., 0.],
-       [0., 0., 0., 0.]])
->>> bm_arr = bm.zeros((2, 4));  bm_arr
-JaxArray([[0., 0., 0., 0.],
-          [0., 0., 0., 0.]], dtype=float32)
-
-# in-place updating
->>> np_arr[0] += 1.;  np_arr
-array([[1., 1., 1., 1.],
-       [0., 0., 0., 0.]])
->>> bm_arr[0] += 1.;  bm_arr
-JaxArray([[1., 1., 1., 1.],
-          [0., 0., 0., 0.]], dtype=float32)
-
-# random number generation
->>> np.random.uniform(-0.1, 0.1, (2, 3))
-array([[-0.02773637,  0.03766689, -0.01363128],
-       [-0.01946991, -0.06669802,  0.09426067]])
->>> bm.random.uniform(-0.1, 0.1, (2, 3))
-JaxArray([[-0.03044081, -0.07787752,  0.04346445],
-          [-0.01366713, -0.0522548 ,  0.04372055]], dtype=float32)
-```
-
-
-
-### 2. Integrator level
-
-Numerical methods for ordinary differential equations (ODEs). 
-
-```python
-sigma = 10; beta = 8/3; rho = 28
-
-@bp.odeint(method='rk4')
-def lorenz_system(x, y, z, t):
-    dx = sigma * (y - x)
-    dy = x * (rho - z) - y
-    dz = x * y - beta * z
-    return dx, dy, dz
-```
-
-
-
-Numerical methods for stochastic differential equations (SDEs). 
-
-```python
-sigma = 10; beta = 8/3; rho = 28
-
-def lorenz_noise(x, y, z, t):
-    return 0.1*x, 0.1*y, 0.1*z
-
-@bp.odeint(method='milstein', g=lorenz_noise)
-def lorenz_system(x, y, z, t):
-    dx = sigma * (y - x)
-    dy = x * (rho - z) - y
-    dz = x * y - beta * z
-    return dx, dy, dz
-```
-
-
-
-Numerical methods for delay differential equations (DDEs).
-
-```python
-xdelay = bm.TimeDelay(bm.zeros(1), delay_len=1., before_t0=1., dt=0.01)
-
-@bp.ddeint(method='rk4', state_delays={'x': xdelay})
-def second_order_eq(x, y, t):
-  dx = y
-  dy = -y - 2 * x - 0.5 * xdelay(t - 1)
-  return dx, dy
-```
-
-
-Numerical methods for fractional differential equations (FDEs).
-
-```python
-sigma = 10; beta = 8/3; rho = 28
-
-@bp.fdeint(method='GLShortMemory', alpha=0.97)
-def fractional_lorenz(x, y, z, t):
-    dx = sigma * (y - x)
-    dy = x * (rho - z) - y
-    dz = x * y - beta * z
-    return dx, dy, dz
-```
-
-
-### 3. Dynamics simulation level
-
-Building an E-I balance network.
-
-```python
-class EINet(bp.dyn.Network):
-  def __init__(self):
-    E = bp.dyn.LIF(3200, V_rest=-60., V_th=-50., V_reset=-60., tau=20., tau_ref=5.)
-    I = bp.dyn.LIF(800, V_rest=-60., V_th=-50., V_reset=-60., tau=20., tau_ref=5.)
-    E.V[:] = bp.math.random.randn(3200) * 2 - 60.
-    I.V[:] = bp.math.random.randn(800) * 2 - 60.
-
-    E2E = bp.dyn.ExpCOBA(E, E, bp.conn.FixedProb(prob=0.02), E=0., g_max=0.6, tau=5.)
-    E2I = bp.dyn.ExpCOBA(E, I, bp.conn.FixedProb(prob=0.02), E=0., g_max=0.6, tau=5.)
-    I2E = bp.dyn.ExpCOBA(I, E, bp.conn.FixedProb(prob=0.02), E=-80., g_max=6.7, tau=10.)
-    I2I = bp.dyn.ExpCOBA(I, I, bp.conn.FixedProb(prob=0.02), E=-80., g_max=6.7, tau=10.)
-
-    super(EINet, self).__init__(E2E, E2I, I2E, I2I, E=E, I=I)
-```
-
-Simulating a whole-brain network by using rate models.
-
-```python
-class WholeBrainNet(bp.dyn.Network):
-  def __init__(self):
-    super(WholeBrainNet, self).__init__()
-
-    self.areas = bp.dyn.RateFHN(80, x_ou_sigma=0.01, y_ou_sigma=0.01, name='fhn')
-    self.conns = bp.dyn.DiffusiveDelayCoupling(self.areas, self.areas, 'x->input',
-                                               conn_mat=conn_mat, 
-                                               delay_mat=delay_mat)
-
-  def update(self, _t, _dt):
-    self.conns.update(_t, _dt)
-    self.areas.update(_t, _dt)
-```
-
-
-
-### 4. Dynamics training level
-
-Training an  echo state network.
-
-```python
-i = bp.nn.Input(3)
-r = bp.nn.Reservoir(100)
-o = bp.nn.LinearReadout(3)
-
-net = i >> r >> o
-
-trainer = bp.nn.RidgeTrainer(net, beta=1e-5)  # Ridge Regression
-
-trainer = bp.nn.ForceTrainer(net, alpha=1.)  # FORCE Learning
-```
-
-
-
-Training a next-generation reservoir computing model.
-
-```python
-i = bp.nn.Input(3)
-r = bp.nn.NVAR(delay=2, order=2)
-o = bp.nn.LinearReadout(3)
-
-net = i >> r >> o
-
-trainer = bp.nn.RidgeTrainer(net, beta=1e-5)
-```
-
-
-
-Training an artificial recurrent neural network. 
-
-```python
-i = bp.nn.Input(3)
-l1 = bp.nn.VanillaRNN(100)
-l2 = bp.nn.VanillaRNN(200)
-o = bp.nn.Dense(10)
-
-net = i >> l1 >> l2 >> o
-
-trainer = bp.nn.BPTT(net, 
-                     loss='cross_entropy_loss',
-                     optimizer=bp.optim.Adam(0.01))
-```
-
-
-
-### 5. Dynamics analysis level
-
-Analyzing a low-dimensional FitzHugh–Nagumo neuron model.
-
-```python
-model = bp.dyn.FHN(1)
-
-analyzer = bp.analysis.PhasePlane2D(
-   model,
-   target_vars={'V': [-3, 3], 'w': [-3., 3.]},
-   pars_update={'I_ext': 0.8}, 
-   resolutions=0.01
-)
-analyzer.plot_nullcline()
-analyzer.plot_vector_field()
-analyzer.plot_fixed_point()
-analyzer.show_figure()
-```
-
-Analyzing a high-dimensional continuous-attractor neural network (CANN).
-
-```python
-cann_model = CANN(100) # your high-dimensional CANN network
-
-finder = bp.analysis.SlowPointFinder(f_cell=cann_model)
-finder.find_fps_with_gd_method(candidates=bm.random.random((1000, 100)))
-finder.filter_loss(tolerance=1e-5)
-finder.keep_unique(tolerance=0.03)
-finder.exclude_outliers(0.1)
-```
-
-
-### 6. More others
-
-For **more functions and examples**, please refer to the [documentation](https://brainpy.readthedocs.io/) and [examples](https://brainpy-examples.readthedocs.io/).
-
-
 ## License
 
 [GNU General Public License v3.0](https://github.com/PKU-NIP-Lab/BrainPy/blob/master/LICENSE)

brainpy/__init__.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-__version__ = "2.1.5"
+__version__ = "2.1.7"
 
 
 try:

brainpy/dyn/synapses/abstract_models.py

@@ -1227,8 +1227,9 @@ def update(self, _t, _dt):
         post_g = bm.sum(self.g)
         if not self.conn.include_self:
           post_g = post_g - self.g
+        post_g = post_g * self.g_max
       else:
-        post_g = self.g * self.g_max
+        post_g = self.g @ self.g_max
     elif isinstance(self.conn, One2One):
       post_g = self.g_max * self.g
     else:
@@ -1242,4 +1243,4 @@ def update(self, _t, _dt):
 
     # output
     g_inf = 1 + self.cc_Mg / self.beta * bm.exp(-self.alpha * self.post.V)
-    self.post.input -= post_g * (self.post.V - self.E) / g_inf
+    self.post.input += post_g * (self.E - self.post.V) / g_inf

changelog.rst

@@ -6,6 +6,73 @@ brainpy 2.x (LTS)
 *****************
 
 
+Version 2.1.6 (2022.04.20)
+==========================
+
+
+What's Changed
+~~~~~~~~~~~~~~
+
+
+
+
+Version 2.1.5 (2022.04.18)
+==========================
+
+
+What's Changed
+~~~~~~~~~~~~~~
+
+* ``brainpy.math.random.shuffle`` is numpy like by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#153 <https://github.com/PKU-NIP-Lab/BrainPy/pull/153>`_
+* update LICENSE by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#155 <https://github.com/PKU-NIP-Lab/BrainPy/pull/155>`_
+* docs: add m1 warning by `@ztqakita <https://github.com/ztqakita>`_ in `#154 <https://github.com/PKU-NIP-Lab/BrainPy/pull/154>`_
+* compatible apis of 'brainpy.math' with those of 'jax.numpy' in most modules by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#156 <https://github.com/PKU-NIP-Lab/BrainPy/pull/156>`_
+* Important updates by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#157 <https://github.com/PKU-NIP-Lab/BrainPy/pull/157>`_
+* Updates by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#159 <https://github.com/PKU-NIP-Lab/BrainPy/pull/159>`_
+* Add LayerNorm, GroupNorm, and InstanceNorm as nn_nodes in normalization.py by `@c-xy17 <https://github.com/c-xy17>`_ in `#162 <https://github.com/PKU-NIP-Lab/BrainPy/pull/162>`_
+* feat: add conv & pooling nodes by `@ztqakita <https://github.com/ztqakita>`_ in `#161 <https://github.com/PKU-NIP-Lab/BrainPy/pull/161>`_
+* fix: update setup.py by `@ztqakita <https://github.com/ztqakita>`_ in `#163 <https://github.com/PKU-NIP-Lab/BrainPy/pull/163>`_
+* update setup.py by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#165 <https://github.com/PKU-NIP-Lab/BrainPy/pull/165>`_
+* fix: change trigger condition by `@ztqakita <https://github.com/ztqakita>`_ in `#166 <https://github.com/PKU-NIP-Lab/BrainPy/pull/166>`_
+* fix: add build_conn() function by `@ztqakita <https://github.com/ztqakita>`_ in `#164 <https://github.com/PKU-NIP-Lab/BrainPy/pull/164>`_
+* update synapses by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#167 <https://github.com/PKU-NIP-Lab/BrainPy/pull/167>`_
+* get the deserved name: brainpy by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#168 <https://github.com/PKU-NIP-Lab/BrainPy/pull/168>`_
+* update tests by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#169 <https://github.com/PKU-NIP-Lab/BrainPy/pull/169>`_
+
+**Full Changelog**\ : `V2.1.4...V2.1.5 <https://github.com/PKU-NIP-Lab/BrainPy/compare/V2.1.4...V2.1.5>`_
+
+
+
+Version 2.1.4 (2022.04.04)
+==========================
+
+
+What's Changed
+~~~~~~~~~~~~~~
+
+* fix doc parsing bug by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#127 <https://github.com/PKU-NIP-Lab/BrainPy/pull/127>`_
+* Update overview_of_dynamic_model.ipynb by `@c-xy17 <https://github.com/c-xy17>`_ in `#129 <https://github.com/PKU-NIP-Lab/BrainPy/pull/129>`_
+* Reorganization of ``brainpylib.custom_op`` and adding interface in ``brainpy.math`` by `@ztqakita <https://github.com/ztqakita>`_ in `#128 <https://github.com/PKU-NIP-Lab/BrainPy/pull/128>`_
+* Fix: modify ``register_op`` and brainpy.math interface by `@ztqakita <https://github.com/ztqakita>`_ in `#130 <https://github.com/PKU-NIP-Lab/BrainPy/pull/130>`_
+* new features about RNN training and delay differential equations by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#132 <https://github.com/PKU-NIP-Lab/BrainPy/pull/132>`_
+* Fix `#123 <https://github.com/PKU-NIP-Lab/BrainPy/issues/123>`_\ : Add low-level operators docs and modify register_op by `@ztqakita <https://github.com/ztqakita>`_ in `#134 <https://github.com/PKU-NIP-Lab/BrainPy/pull/134>`_
+* feat: add generate_changelog by `@ztqakita <https://github.com/ztqakita>`_ in `#135 <https://github.com/PKU-NIP-Lab/BrainPy/pull/135>`_
+* fix `#133 <https://github.com/PKU-NIP-Lab/BrainPy/issues/133>`_\ , support batch size training with offline algorithms by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#136 <https://github.com/PKU-NIP-Lab/BrainPy/pull/136>`_
+* fix `#84 <https://github.com/PKU-NIP-Lab/BrainPy/issues/84>`_\ : support online training algorithms by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#137 <https://github.com/PKU-NIP-Lab/BrainPy/pull/137>`_
+* feat: add the batch normalization node by `@c-xy17 <https://github.com/c-xy17>`_ in `#138 <https://github.com/PKU-NIP-Lab/BrainPy/pull/138>`_
+* fix: fix shape checking error by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#139 <https://github.com/PKU-NIP-Lab/BrainPy/pull/139>`_
+* solve `#131 <https://github.com/PKU-NIP-Lab/BrainPy/issues/131>`_\ , support efficient synaptic computation for special connection types by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#140 <https://github.com/PKU-NIP-Lab/BrainPy/pull/140>`_
+* feat: update the API and test for batch normalization by `@c-xy17 <https://github.com/c-xy17>`_ in `#142 <https://github.com/PKU-NIP-Lab/BrainPy/pull/142>`_
+* Node is default trainable by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#143 <https://github.com/PKU-NIP-Lab/BrainPy/pull/143>`_
+* Updates training apis and docs by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#145 <https://github.com/PKU-NIP-Lab/BrainPy/pull/145>`_
+* fix: add dependencies and update version by `@ztqakita <https://github.com/ztqakita>`_ in `#147 <https://github.com/PKU-NIP-Lab/BrainPy/pull/147>`_
+* update requirements by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#146 <https://github.com/PKU-NIP-Lab/BrainPy/pull/146>`_
+* data pass of the Node is default SingleData by `@chaoming0625 <https://github.com/chaoming0625>`_ in `#148 <https://github.com/PKU-NIP-Lab/BrainPy/pull/148>`_
+
+**Full Changelog**\ : `V2.1.3...V2.1.4 <https://github.com/PKU-NIP-Lab/BrainPy/compare/V2.1.3...V2.1.4>`_
+
+
+
 Version 2.1.3 (2022.03.27)
 ==========================