Classes
class	AvgPool1D

class	BatchNorm1D

struct	BatchNormParams
	Parameters for BatchNorm1D. More...

class	Conv1D

struct	Conv1DParams
	Parameters for a Conv1D layer. More...

class	Dense
	Fully-connected layer: `Z = W*X + b`, `A = act(Z)`. More...

struct	DenseParams
	Parameters for a Dense layer. More...

class	Layer
	Type-erased layer — holds one concrete layer type. More...

class	MaxPool1D

struct	Pool1DParams
	Parameters for pooling layers (Max and Avg share this POD). More...

Enumerations
enum class	LayerType : uint8_t { Dense = 0 , Conv1D = 1 , BatchNorm1D = 2 , MaxPool1D = 3 , AvgPool1D = 4 }
	Layer kind, stable on-wire (uint8_t). More...

enum class	DType : uint16_t { F32 = 0 }
	Data type for blobs in `model.bin`. More...

Enumeration Type Documentation

◆ DType

enum class FlexNN::Layers::DType : uint16_t

strong

Data type for blobs in model.bin.

v0.1 only F32 (float32 IEEE). Value 1 is reserved; importer must reject unknown values (e.g., !=0) with Status error.

Enumerator
F32	float32 little-endian

strong

Layer kind, stable on-wire (uint8_t).

Do not reorder — model.bin stores this as 1 byte. New types are appended.

Enumerator
Dense	Fully-connected: W[out×in], b[out].
Conv1D	1D temporal convolution
BatchNorm1D	Per-feature BatchNorm (gamma/beta/mean/var in aux)
MaxPool1D	Max pooling, weightless.
AvgPool1D	Average pooling, weightless.