SimplePCNetwork.h File Reference

Network wrapper for SimplePCLayer, mirroring DiscriminativePCNetwork exactly (minus the UpdatePrecision() call, which no longer exists). More...

#include <vector>
#include <memory>
#include <random>
#include <deepity/layers/SimplePCLayer.h>
#include <deepity/utils/MemoryArena.h>
#include <deepity/utils/DeviceMemoryArena.h>
#include <deepity/backend/Backend.h>

Go to the source code of this file.

Classes
class	Deep::SimplePCNetwork
	An abstracted class for an array of SimplePCLayer. More...

Namespaces
namespace	Deep
	This file exists to provide both Tensor.h and IComputeBackend.h with device types without introducing circular dependencies.

Detailed Description

Network wrapper for SimplePCLayer, mirroring DiscriminativePCNetwork exactly (minus the UpdatePrecision() call, which no longer exists).

This header includes implementations of PC layer-to-layer interaction.

#include <deepity/networks/SimplePCNetwork.h>
 
Deep::SimplePCNetwork network(1);
network.addLayer({...});
network.Clamp(input);
network.CalculateState();

Note

All layers are stored in a vector.

As of this revision, the network owns a single IComputeBackend (CPU by default, GPU if requested at construction) and passes it down into every layer it creates. device defaults to DEVICE_CPU, so every existing caller (nanobind bindings, hand-written C++) keeps compiling and behaving exactly as before – passing DEVICE_GPU explicitly is only needed for new, GPU-aware call sites.

Version

1.1

Date

2026-09-05

Author

Jack Rose