Pytorch 2.0

.github/workflows/python-app.yml

@@ -22,7 +22,7 @@ jobs:
         python-version: 3.9
     - name: Install Poetry
       env:
-        POETRY_VERSION: 1.1.13
+        POETRY_VERSION: 1.4.2
       run: |
         curl -sSL https://raw.githubusercontent.com/python-poetry/poetry/master/install-poetry.py | python - &&\
             poetry config virtualenvs.create false

.github/workflows/pythonpackage.yml

@@ -19,7 +19,7 @@ jobs:
         python-version: ${{ matrix.python-version }}
     - name: Install Poetry
       env:
-        POETRY_VERSION: 1.1.13
+        POETRY_VERSION: 1.4.2
       run: |
         curl -sSL https://raw.githubusercontent.com/python-poetry/poetry/master/install-poetry.py | python - &&\
             poetry config virtualenvs.create false

bindsnet/datasets/collate.py

@@ -10,7 +10,6 @@
 import collections
 
 import torch
-from torch._six import string_classes
 from torch.utils.data._utils import collate as pytorch_collate
 
 
@@ -75,8 +74,6 @@ def time_aware_collate(batch):
         return torch.tensor(batch, dtype=torch.float64)
     elif isinstance(elem, int):
         return torch.tensor(batch)
-    elif isinstance(elem, string_classes):
-        return batch
     elif isinstance(elem, collections.Mapping):
         return {key: time_aware_collate([d[key] for d in batch]) for key in elem}
     elif isinstance(elem, tuple) and hasattr(elem, "_fields"):  # namedtuple

bindsnet/pipeline/base_pipeline.py

@@ -3,8 +3,6 @@
 from typing import Any, Dict, Tuple
 
 import torch
-from torch._six import string_classes
-
 from bindsnet.network import Network
 from bindsnet.network.monitors import Monitor
 
@@ -23,7 +21,7 @@ def recursive_to(item, device):
 
     if isinstance(item, torch.Tensor):
         return item.to(device)
-    elif isinstance(item, (string_classes, int, float, bool)):
+    elif isinstance(item, (int, float, bool)):
         return item
     elif isinstance(item, collections.abc.Mapping):
         return {key: recursive_to(item[key], device) for key in item}

examples/mnist/reservoir.py

@@ -140,7 +140,7 @@
 
     # Run network on sample image
     network.run(inputs={"I": datum}, time=time)
-    training_pairs.append([spikes["O"].get("s").sum(0), label])
+    training_pairs.append([spikes["O"].get("s"), label])
 
     # Plot spiking activity using monitors
     if plot:
@@ -187,7 +187,7 @@ def forward(self, x):
 
 
 # Create and train logistic regression model on reservoir outputs.
-model = NN(n_neurons, 10).to(device)
+model = NN(n_neurons * args.time, 10).to(device)
 criterion = torch.nn.MSELoss(reduction="sum")
 optimizer = torch.optim.SGD(model.parameters(), lr=1e-4, momentum=0.9)
 

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "bindsnet"
-version = "0.3.1"
+version = "0.3.2"
 description = "Spiking neural networks for ML in Python"
 authors = [ "Hananel Hazan <hananel@hazan.org.il>", "Daniel Saunders", "Darpan Sanghavi", "Hassaan Khan" ]
 license = "AGPL-3.0-only"
@@ -14,13 +14,13 @@ python = ">=3.8,<3.11"
 numpy = "^1.24.2"
 scipy = "^1.9.1"
 Cython = "^0.29.33"
-torch = "1.13.1"
-torchvision = "0.14.1"
-torchaudio = "0.13.1"
+torch = "2.0.0"
+torchvision = "0.15.1"
+torchaudio = "2.0.1"
 tensorboardX = "2.6"
 tqdm = "^4.65.0"
 matplotlib = "^3.7.1"
-gymnasium = "^0.27.1"
+gymnasium = "^0.28.1"
 scikit-build = "^0.16.7"
 scikit-image = "^0.20.0"
 scikit-learn = "^1.2.1"