peal.architectures

Predictor architectures and the building blocks they are assembled from.

interfaces declares TaskConfig / ArchitectureConfig and the per-family configs (FC, VGG, ResNet, Transformer); predictors builds and loads the models PEAL trains and explains (SequentialModel, TorchvisionModel); module_blocks holds the VGG / ResNet / Transformer blocks they are made of; basic_modules contains small tensor-shaping layers and onnx_predictor wraps exported ONNX classifiers so they can be explained like native nn.Module predictors.

Submodules

basic_modules	Small nn.Module building blocks used by PEAL's configurable architectures.
interfaces	Pydantic config templates for PEAL architectures and training tasks.
module_blocks	Building blocks from which PEAL's configurable predictors are assembled.
onnx_predictor	ONNX classifiers as PEAL predictors.
predictors	Predictor architectures and the loaders that turn a config into a model.