Models¶
Pre-built network architectures and model registry.
SCDigitClassifier— Pre-configured SC network for MNIST digit classification. Architecture: Conv (28x28, 1ch->4ch, kernel 3, stride 2) + Vectorized dense + output.SCIzhikevichNeuron— maintained Izhikevich software neuron reference with explicit baseline/RK4 integrator paths, cross-language RK4 parity, and latest benchmark evidence.
10 pre-built configurations in the model zoo:
| Config | Task |
|---|---|
| Brunel balanced | E/I balance dynamics |
| Cortical column | Layered cortical model |
| CPG | Central pattern generator |
| Decision-making | 2-pool WTA |
| Working memory | Persistent activity |
| Visual cortex V1 | Orientation selectivity |
| MNIST classifier | Digit recognition |
| SHD classifier | Speech (Spiking Heidelberg Digits) |
| DVS gesture | Event camera gestures |
| Auditory | Sound processing |
Python
from sc_neurocore.models import SCDigitClassifier
model = SCDigitClassifier()
output = model.forward(image_28x28)
sc_neurocore.models.zoo
¶
SCDigitClassifier
¶
Pre-configured SC Network for MNIST-like Digit Classification. Uses: Conv Layer -> Vectorized Dense Layer
Source code in src/sc_neurocore/models/zoo.py
| Python | |
|---|---|
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 | |
forward(image)
¶
Classify a 28x28 image.
Source code in src/sc_neurocore/models/zoo.py
| Python | |
|---|---|
33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 | |
SCKeywordSpotter
¶
Audio Keyword Spotter (e.g., "Yes"/"No"). Uses: Recurrent / Dense Layer
Source code in src/sc_neurocore/models/zoo.py
| Python | |
|---|---|
62 63 64 65 66 67 68 69 70 71 72 73 | |