merge with master

.gitignore

@@ -18,6 +18,7 @@ test/models/__pycache__/*
 test/network/__pycache__/*
 test/analysis/__pycache__/*
 *.pyc
+**/*.pyc
 dist/*
 logs/*
 .pytest_cache/*

README.md

@@ -9,9 +9,9 @@ This package is used as part of ongoing research on applying SNNs, machine learn
 
 Check out the [BindsNET examples](https://github.com/BindsNET/bindsnet/tree/master/examples) for a collection of experiments, functions for the analysis of results, plots of experiment outcomes, and more. Documentation for the package can be found [here](https://bindsnet-docs.readthedocs.io).
 
-![Build Status](https://github.com/BindsNET/bindsnet/actions/workflows/python-app.yml/badge.svg?branch=master)
+[![CodeQL](https://github.com/BindsNET/bindsnet/actions/workflows/github-code-scanning/codeql/badge.svg)](https://github.com/BindsNET/bindsnet/actions/workflows/github-code-scanning/codeql)
 [![Documentation Status](https://readthedocs.org/projects/bindsnet-docs/badge/?version=latest)](https://bindsnet-docs.readthedocs.io/?badge=latest)
-[![Gitter chat](https://badges.gitter.im/gitterHQ/gitter.png)](https://gitter.im/bindsnet_/community)
+[![Neuromorphic Computing](https://img.shields.io/badge/Collaboration_Network-Open_Neuromorphic-blue)](https://open-neuromorphic.org/neuromorphic-computing/)
 
 ## Requirements
 

bindsnet/evaluation/evaluation.py

@@ -44,8 +44,11 @@ def assign_labels(
             indices = torch.nonzero(labels == i).view(-1)
 
             # Compute average firing rates for this label.
+            selected_spikes = torch.index_select(
+                spikes, dim=0, index=torch.tensor(indices)
+            )
             rates[:, i] = alpha * rates[:, i] + (
-                torch.sum(spikes[indices], 0) / n_labeled
+                torch.sum(selected_spikes, 0) / n_labeled
             )
 
     # Compute proportions of spike activity per class.
@@ -111,6 +114,8 @@ def all_activity(
 
     # Sum over time dimension (spike ordering doesn't matter).
     spikes = spikes.sum(1)
+    if spikes.is_sparse:
+        spikes = spikes.to_dense()
 
     rates = torch.zeros((n_samples, n_labels), device=spikes.device)
     for i in range(n_labels):
@@ -152,6 +157,8 @@ def proportion_weighting(
 
     # Sum over time dimension (spike ordering doesn't matter).
     spikes = spikes.sum(1)
+    if spikes.is_sparse:
+        spikes = spikes.to_dense()
 
     rates = torch.zeros((n_samples, n_labels), device=spikes.device)
     for i in range(n_labels):

bindsnet/learning/MCC_learning.py

@@ -102,7 +102,12 @@ def update(self, **kwargs) -> None:
         if ((self.min is not None) or (self.max is not None)) and not isinstance(
             self, NoOp
         ):
-            self.feature_value.clamp_(self.min, self.max)
+            if self.feature_value.is_sparse:
+                self.feature_value = (
+                    self.feature_value.to_dense().clamp_(self.min, self.max).to_sparse()
+                )
+            else:
+                self.feature_value.clamp_(self.min, self.max)
 
     @abstractmethod
     def reset_state_variables(self) -> None:
@@ -247,10 +252,15 @@ def _connection_update(self, **kwargs) -> None:
                         torch.mean(self.average_buffer_pre, dim=0) * self.connection.dt
                     )
             else:
-                self.feature_value -= (
-                    self.reduction(torch.bmm(source_s, target_x), dim=0)
-                    * self.connection.dt
-                )
+                if self.feature_value.is_sparse:
+                    self.feature_value -= (
+                        torch.bmm(source_s, target_x) * self.connection.dt
+                    ).to_sparse()
+                else:
+                    self.feature_value -= (
+                        self.reduction(torch.bmm(source_s, target_x), dim=0)
+                        * self.connection.dt
+                    )
             del source_s, target_x
 
         # Post-synaptic update.
@@ -278,10 +288,15 @@ def _connection_update(self, **kwargs) -> None:
                         torch.mean(self.average_buffer_post, dim=0) * self.connection.dt
                     )
             else:
-                self.feature_value += (
-                    self.reduction(torch.bmm(source_x, target_s), dim=0)
-                    * self.connection.dt
-                )
+                if self.feature_value.is_sparse:
+                    self.feature_value += (
+                        torch.bmm(source_x, target_s) * self.connection.dt
+                    ).to_sparse()
+                else:
+                    self.feature_value += (
+                        self.reduction(torch.bmm(source_x, target_s), dim=0)
+                        * self.connection.dt
+                    )
             del source_x, target_s
 
         super().update()
@@ -508,16 +523,16 @@ def _connection_update(self, **kwargs) -> None:
                 self.average_buffer_index + 1
             ) % self.average_update
 
-            if self.continues_update:
-                self.feature_value += self.nu[0] * torch.mean(
-                    self.average_buffer, dim=0
-                )
-            elif self.average_buffer_index == 0:
-                self.feature_value += self.nu[0] * torch.mean(
-                    self.average_buffer, dim=0
-                )
+            if self.continues_update or self.average_buffer_index == 0:
+                update = self.nu[0] * torch.mean(self.average_buffer, dim=0)
+                if self.feature_value.is_sparse:
+                    update = update.to_sparse()
+                self.feature_value += update
         else:
-            self.feature_value += self.nu[0] * self.reduction(update, dim=0)
+            update = self.nu[0] * self.reduction(update, dim=0)
+            if self.feature_value.is_sparse:
+                update = update.to_sparse()
+            self.feature_value += update
 
         # Update P^+ and P^- values.
         self.p_plus *= torch.exp(-self.connection.dt / self.tc_plus)
@@ -686,14 +701,16 @@ def _connection_update(self, **kwargs) -> None:
                 self.average_buffer_index + 1
             ) % self.average_update
 
-            if self.continues_update:
-                self.feature_value += torch.mean(self.average_buffer, dim=0)
-            elif self.average_buffer_index == 0:
-                self.feature_value += torch.mean(self.average_buffer, dim=0)
+            if self.continues_update or self.average_buffer_index == 0:
+                update = torch.mean(self.average_buffer, dim=0)
+                if self.feature_value.is_sparse:
+                    update = update.to_sparse()
+                self.feature_value += update
         else:
-            self.feature_value += (
-                self.nu[0] * self.connection.dt * reward * self.eligibility_trace
-            )
+            update = self.nu[0] * self.connection.dt * reward * self.eligibility_trace
+            if self.feature_value.is_sparse:
+                update = update.to_sparse()
+            self.feature_value += update
 
         # Update P^+ and P^- values.
         self.p_plus *= torch.exp(-self.connection.dt / self.tc_plus)  # Decay

bindsnet/learning/learning.py

@@ -98,7 +98,10 @@ def update(self) -> None:
             (self.connection.wmin != -np.inf).any()
             or (self.connection.wmax != np.inf).any()
         ) and not isinstance(self, NoOp):
-            self.connection.w.clamp_(self.connection.wmin, self.connection.wmax)
+            if self.connection.w.is_sparse:
+                raise Exception("SparseConnection isn't supported for wmin\\wmax")
+            else:
+                self.connection.w.clamp_(self.connection.wmin, self.connection.wmax)
 
 
 class NoOp(LearningRule):
@@ -396,7 +399,10 @@ def _connection_update(self, **kwargs) -> None:
         if self.nu[0].any():
             source_s = self.source.s.view(batch_size, -1).unsqueeze(2).float()
             target_x = self.target.x.view(batch_size, -1).unsqueeze(1) * self.nu[0]
-            self.connection.w -= self.reduction(torch.bmm(source_s, target_x), dim=0)
+            update = self.reduction(torch.bmm(source_s, target_x), dim=0)
+            if self.connection.w.is_sparse:
+                update = update.to_sparse()
+            self.connection.w -= update
             del source_s, target_x
 
         # Post-synaptic update.
@@ -405,7 +411,10 @@ def _connection_update(self, **kwargs) -> None:
                 self.target.s.view(batch_size, -1).unsqueeze(1).float() * self.nu[1]
             )
             source_x = self.source.x.view(batch_size, -1).unsqueeze(2)
-            self.connection.w += self.reduction(torch.bmm(source_x, target_s), dim=0)
+            update = self.reduction(torch.bmm(source_x, target_s), dim=0)
+            if self.connection.w.is_sparse:
+                update = update.to_sparse()
+            self.connection.w += update
             del source_x, target_s
 
         super().update()
@@ -1113,10 +1122,14 @@ def _connection_update(self, **kwargs) -> None:
 
         # Pre-synaptic update.
         update = self.reduction(torch.bmm(source_s, target_x), dim=0)
+        if self.connection.w.is_sparse:
+            update = update.to_sparse()
         self.connection.w += self.nu[0] * update
 
         # Post-synaptic update.
         update = self.reduction(torch.bmm(source_x, target_s), dim=0)
+        if self.connection.w.is_sparse:
+            update = update.to_sparse()
         self.connection.w += self.nu[1] * update
 
         super().update()
@@ -1542,8 +1555,10 @@ def _connection_update(self, **kwargs) -> None:
             a_minus = torch.tensor(a_minus, device=self.connection.w.device)
 
         # Compute weight update based on the eligibility value of the past timestep.
-        update = reward * self.eligibility
-        self.connection.w += self.nu[0] * self.reduction(update, dim=0)
+        update = self.reduction(reward * self.eligibility, dim=0)
+        if self.connection.w.is_sparse:
+            update = update.to_sparse()
+        self.connection.w += self.nu[0] * update
 
         # Update P^+ and P^- values.
         self.p_plus *= torch.exp(-self.connection.dt / self.tc_plus)
@@ -2214,10 +2229,11 @@ def _connection_update(self, **kwargs) -> None:
         self.eligibility_trace *= torch.exp(-self.connection.dt / self.tc_e_trace)
         self.eligibility_trace += self.eligibility / self.tc_e_trace
 
+        update = self.nu[0] * self.connection.dt * reward * self.eligibility_trace
+        if self.connection.w.is_sparse:
+            update = update.to_sparse()
         # Compute weight update.
-        self.connection.w += (
-            self.nu[0] * self.connection.dt * reward * self.eligibility_trace
-        )
+        self.connection.w += update
 
         # Update P^+ and P^- values.
         self.p_plus *= torch.exp(-self.connection.dt / self.tc_plus)
@@ -2936,6 +2952,9 @@ def _connection_update(self, **kwargs) -> None:
         ) * source_x[:, None]
 
         # Compute weight update.
-        self.connection.w += self.nu[0] * reward * self.eligibility_trace
+        update = self.nu[0] * reward * self.eligibility_trace
+        if self.connection.w.is_sparse:
+            update = update.to_sparse()
+        self.connection.w += update
 
         super().update()

bindsnet/models/models.py

@@ -4,11 +4,18 @@
 import torch
 from scipy.spatial.distance import euclidean
 from torch.nn.modules.utils import _pair
+from torch import device
 
 from bindsnet.learning import PostPre
+from bindsnet.learning.MCC_learning import PostPre as MMCPostPre
 from bindsnet.network import Network
 from bindsnet.network.nodes import DiehlAndCookNodes, Input, LIFNodes
-from bindsnet.network.topology import Connection, LocalConnection
+from bindsnet.network.topology import (
+    Connection,
+    LocalConnection,
+    MulticompartmentConnection,
+)
+from bindsnet.network.topology_features import Weight
 
 
 class TwoLayerNetwork(Network):
@@ -94,6 +101,9 @@ class DiehlAndCook2015(Network):
     def __init__(
         self,
         n_inpt: int,
+        device: str = "cpu",
+        batch_size: int = None,
+        sparse: bool = False,
         n_neurons: int = 100,
         exc: float = 22.5,
         inh: float = 17.5,
@@ -102,6 +112,7 @@ def __init__(
         reduction: Optional[callable] = None,
         wmin: float = 0.0,
         wmax: float = 1.0,
+        w_dtype: torch.dtype = torch.float32,
         norm: float = 78.4,
         theta_plus: float = 0.05,
         tc_theta_decay: float = 1e7,
@@ -124,6 +135,7 @@ def __init__(
             dimension.
         :param wmin: Minimum allowed weight on input to excitatory synapses.
         :param wmax: Maximum allowed weight on input to excitatory synapses.
+        :param w_dtype: Data type for :code:`w` tensor
         :param norm: Input to excitatory layer connection weights normalization
             constant.
         :param theta_plus: On-spike increment of ``DiehlAndCookNodes`` membrane
@@ -170,27 +182,57 @@ def __init__(
 
         # Connections
         w = 0.3 * torch.rand(self.n_inpt, self.n_neurons)
-        input_exc_conn = Connection(
+        input_exc_conn = MulticompartmentConnection(
             source=input_layer,
             target=exc_layer,
-            w=w,
-            update_rule=PostPre,
-            nu=nu,
-            reduction=reduction,
-            wmin=wmin,
-            wmax=wmax,
-            norm=norm,
+            device=device,
+            pipeline=[
+                Weight(
+                    "weight",
+                    w,
+                    value_dtype=w_dtype,
+                    range=[wmin, wmax],
+                    norm=norm,
+                    reduction=reduction,
+                    nu=nu,
+                    learning_rule=MMCPostPre,
+                    sparse=sparse,
+                    batch_size=batch_size,
+                )
+            ],
         )
         w = self.exc * torch.diag(torch.ones(self.n_neurons))
-        exc_inh_conn = Connection(
-            source=exc_layer, target=inh_layer, w=w, wmin=0, wmax=self.exc
+        if sparse:
+            w = w.unsqueeze(0).expand(batch_size, -1, -1)
+        exc_inh_conn = MulticompartmentConnection(
+            source=exc_layer,
+            target=inh_layer,
+            device=device,
+            pipeline=[
+                Weight(
+                    "weight", w, value_dtype=w_dtype, range=[0, self.exc], sparse=sparse
+                )
+            ],
         )
         w = -self.inh * (
             torch.ones(self.n_neurons, self.n_neurons)
             - torch.diag(torch.ones(self.n_neurons))
         )
-        inh_exc_conn = Connection(
-            source=inh_layer, target=exc_layer, w=w, wmin=-self.inh, wmax=0
+        if sparse:
+            w = w.unsqueeze(0).expand(batch_size, -1, -1)
+        inh_exc_conn = MulticompartmentConnection(
+            source=inh_layer,
+            target=exc_layer,
+            device=device,
+            pipeline=[
+                Weight(
+                    "weight",
+                    w,
+                    value_dtype=w_dtype,
+                    range=[-self.inh, 0],
+                    sparse=sparse,
+                )
+            ],
         )
 
         # Add to network

bindsnet/network/monitors.py

@@ -45,6 +45,7 @@ def __init__(
         time: Optional[int] = None,
         batch_size: int = 1,
         device: str = "cpu",
+        sparse: Optional[bool] = False,
     ):
         # language=rst
         """
@@ -62,6 +63,7 @@ def __init__(
         self.time = time
         self.batch_size = batch_size
         self.device = device
+        self.sparse = sparse
 
         # if time is not specified the monitor variable accumulate the logs
         if self.time is None:
@@ -98,11 +100,12 @@ def record(self) -> None:
         for v in self.state_vars:
             data = getattr(self.obj, v).unsqueeze(0)
             # self.recording[v].append(data.detach().clone().to(self.device))
-            self.recording[v].append(
-                torch.empty_like(data, device=self.device, requires_grad=False).copy_(
-                    data, non_blocking=True
-                )
-            )
+            record = torch.empty_like(
+                data, device=self.device, requires_grad=False
+            ).copy_(data, non_blocking=True)
+            if self.sparse:
+                record = record.to_sparse()
+            self.recording[v].append(record)
             # remove the oldest element (first in the list)
             if self.time is not None:
                 self.recording[v].pop(0)