8f43927575
examples/ is based on KasperskyLab/knp at 55470ed6 with local changes to the mnist-learn altai/blifat models, cuda-backend CMake and model_loader. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
28 lines
No EOL
2.1 KiB
Markdown
28 lines
No EOL
2.1 KiB
Markdown
# Kaspersky Neuromorphic Platform changelog
|
|
|
|
## Release for Kaspersky Neuromorphic Platform 2.0 - 2025-12
|
|
|
|
### Changes
|
|
|
|
* Distribution kit for developing Linux applications: added packages containing a backend library for GPUs with CUDA support technology.
|
|
* Backends:
|
|
* `CUDABackend`: added an interface for the GPU backend.
|
|
* `SingleThreadedCPUBackend` backend: added neuron support based on the AltAILIF model.
|
|
* `MultiThreadedCPUBackend` backend: added neuron support based on the SynapticResourceSTDPBLIFATNeuron model and synapse based on the SynapticResourceSTDPDeltaSynapse model.
|
|
* Optimized CPU backend operation.
|
|
* C++ components:
|
|
* `core` library: extended the functionality for the `Backend`, `MessageBus`, `MessageEndpoint`, `Population`, `Projection`, `Subscription`, and `TagMap` classes.
|
|
* Framework for C ++: extended the functionality of the C++ framework:
|
|
* `data_processing` namespace: added a namespace that provides functionality for processing classification model training and testing data.
|
|
* `inference_evaluation` namespace: added a namespace that provides functionality for evaluating model quality.
|
|
* `modifier` namespace: added a namespace that provides message handlers.
|
|
* `normalization` namespace: added a namespace that provides functions and functors for normalizing the parameters of neurons and synapses in neural networks.
|
|
* `monitoring` namespace: added the model namespace with model monitoring functions.
|
|
* `projection` namespace: added functions to generate synapses and synapse projections.
|
|
* `ModelExecutor` class: added methods.
|
|
* `neuron_traits` library: extended the neuron interfaces for the `BLIFATNeuron` and `SynapticResourceSTDPBLIFATNeuron` models.
|
|
* Components for use in Python: added methods for the `ModelExecutor` and `UID` classes.
|
|
* `synapse_traits` library (for use in C++ and Python): extended the interface of the synapse based on the `SynapticResourceSTDPDeltaSynapse` model.
|
|
* Example `mnist-learn` is an added example of training a neural network on images of handwritten numbers and their labels from the MNIST database and its execution.
|
|
|
|
© 2025 AO Kaspersky Lab |