This inheritance list is sorted roughly, but not completely, alphabetically:

[detail level 123]

C__half
C__nv_bfloat16
Copennn::AugmentationSettings	Image augmentation parameters: reflections, rotations and translations applied at training
Copennn::Backend	Process-wide singleton that owns the thread pool and the cuBLAS/cuDNN handles
Copennn::BackPropagation	Workspace holding parameter gradients and per-layer deltas during a backward pass
Copennn::BackPropagationLM	Backprop scratch state specific to Levenberg-Marquardt (per-sample errors, Jacobian, Hessian approx)
Copennn::Batch	Minibatch container holding pinned host/device buffers and views into a Dataset
Copennn::TestingAnalysis::BinaryClassificationRates	Sample indices split into the four cells of a binary classification confusion matrix
Copennn::BoxPlot	Five-number summary (minimum, Q1, median, Q3, maximum) used to draw a box plot
Copennn::Buffer	Owning raw byte buffer that lives on CPU or CUDA memory, with aligned (re)allocation
Copennn::ResponseOptimization::Condition	Constraint or objective imposed on a single variable, with optional bounds
Copennn::CsvReader::Config	Reader configuration: field separator and an optional per-line validator
Copennn::Configuration	Global singleton holding the OpenNN device and precision configuration
Copennn::Correlation	Result of a correlation analysis: model parameters, fit quality, and the method/form used
Copennn::CsvReader	Tokenising CSV reader that returns string_views into a single backing buffer
CcudnnTensorStruct
►Copennn::Dataset	Abstract base class for OpenNN datasets, owning samples, variables, and metadata
Copennn::ImageDataset	Image dataset that streams BMP images from disk with optional augmentation
Copennn::LanguageDataset	Token-based language dataset with input/target vocabularies and binary token cache
►Copennn::TabularDataset	Tabular dataset with CSV loading, scaling, descriptive statistics and correlation analysis
Copennn::TimeSeriesDataset	Time series dataset with configurable past/future windows and autocorrelation analysis
Copennn::Descriptives	Summary statistics (minimum, maximum, mean, standard deviation) for one variable
Copennn::ResponseOptimization::Domain	Bounded domain in input or output space defined by inferior and superior frontiers
Copennn::Stats::Entry
Copennn::EnumMap< Enum >
Copennn::Optimizer::EpochStats	Aggregated per-epoch error and accuracy returned by training/evaluation passes
Copennn::Loss::EvaluationResult	Result of calculate_error; accuracy and active_tokens_count are populated only by classification losses
Copennn::FileReader	Thread-safe positional file reader (pread on POSIX, overlapped ReadFile on Windows)
Copennn::FileWriter	Streaming writer that finalises by atomic-renaming a .tmp file to its final path
Copennn::ForwardPropagation	Workspace holding the activations of every layer during a forward pass
Copennn::TestingAnalysis::GoodnessOfFitAnalysis	Coefficient of determination and the matching target/output series for a single output variable
Copennn::Histogram	Frequency histogram with per-bin minimums, maximums, centers, and counts
►Copennn::InputsSelection	Abstract base class for algorithms that search the optimal subset of input variables
Copennn::GeneticAlgorithm	Selects the optimal subset of input features using an evolutionary genetic algorithm
Copennn::GrowingInputs	Selects the optimal subset of input features by greedily growing the input set
Copennn::InputsSelectionResults	Aggregated results of an inputs selection run including optimal inputs and error histories
Copennn::Json
Copennn::JsonDocument
Copennn::JsonWriter
Copennn::KMeans	K-means clustering utility that partitions samples into the requested number of clusters
Copennn::TestingAnalysis::KolmogorovSmirnovResults	Results of a Kolmogorov-Smirnov analysis: cumulative gains and maximum gain
►Copennn::Layer	Abstract base class for all OpenNN layers; orchestrates operators and shape propagation
Copennn::Addition	Element-wise addition layer that sums two input tensors of identical shape (residual connections)
Copennn::Bounding	Output bounding layer that clips outputs to configured lower and upper limits
Copennn::Convolutional	2D convolutional layer with kernel, stride, padding, activation, and optional batch normalization
Copennn::ConvolutionalRelu	Fused convolution + ReLU layer; runs as a single GPU op (cudnn) and is CUDA-Graph friendly
Copennn::Dense	Fully-connected layer with configurable activation, optional batch normalization and dropout
Copennn::DenseRelu	Fused dense + ReLU layer; combines linear projection and ReLU activation in a single op for performance
Copennn::Embedding	Token-id to dense vector embedding layer with optional scaling and positional encoding
Copennn::Flatten	Flatten layer that reshapes a multi-dimensional input into a single 1D feature vector
Copennn::MultiHeadAttention	Multi-head scaled dot-product attention layer used in transformer architectures
Copennn::Normalization3d	Layer normalization over the embedding axis of a 3D (batch, sequence, embedding) tensor
Copennn::Pooling	2D spatial pooling layer supporting max and average reduction
Copennn::Pooling3d	Sequence pooling layer reducing the time axis of a (sequence, features) input
Copennn::Recurrent	Basic recurrent (RNN) layer that unrolls over time steps with a shared activation
Copennn::Scaling	Input scaling layer that normalizes features using per-variable descriptive statistics
Copennn::Unscaling	Output unscaling layer that reverts normalization back to the original feature ranges
Copennn::Loss	Unified loss container supporting MSE, cross-entropy, Minkowski, weighted, and regularized variants
Copennn::ModelExpression	Emits a trained neural network as source code in C, Python, JavaScript, or PHP
Copennn::ModelSelection	Orchestrates model selection by combining inputs selection and neurons selection over a TrainingStrategy
►Copennn::NeuralNetwork	Container of layers forming a feed-forward neural network, with parameter storage and I/O
Copennn::ApproximationNetwork	Factory neural network preconfigured for regression / function approximation
Copennn::AutoAssociationNetwork	Factory neural network preconfigured for auto-association (anomaly detection)
Copennn::ClassificationNetwork	Factory neural network preconfigured for tabular classification
Copennn::ForecastingNetwork	Factory neural network preconfigured for time-series forecasting
Copennn::ImageClassificationNetwork	Factory convolutional neural network preconfigured for image classification
Copennn::SimpleResNet	Factory residual neural network with a configurable number of blocks per stage
Copennn::TextClassificationNetwork	Factory neural network preconfigured for text classification
Copennn::Transformer	Factory encoder-decoder Transformer neural network for sequence-to-sequence tasks
Copennn::VGG16	Factory neural network reproducing the VGG-16 architecture
►Copennn::NeuronSelection	Abstract base class for algorithms that select the optimal number of hidden neurons
Copennn::GrowingNeurons	Selects the optimal hidden neuron count by incrementally growing the number of neurons
Copennn::NeuronsSelectionResults	Aggregated results of a neurons selection run including the optimal neuron count and error histories
Copennn::ResponseOptimization::Objectives	Encodes the objectives extracted from the response optimization configuration
►Copennn::Operator	Base class for compute building blocks composed by layers (matmul, activation, dropout, etc.)
Copennn::ActivationOp	Element-wise non-linear activation (Identity, Sigmoid, Tanh, ReLU, Softmax)
Copennn::AddOp	Element-wise sum of several input tensors (used by residual connections)
Copennn::AttentionOp	Scaled dot-product attention with optional causal mask and dropout
Copennn::BatchNormOp	Batch normalization with learnable scale/shift and running statistics for inference
Copennn::BoundOp	Clamps each output channel to a configurable lower/upper interval
Copennn::CombinationOp	Affine combination output = input * weights + bias (the dense matmul building block)
Copennn::CombinationReluOp	Fused affine + ReLU activation (uses cuBLASLt epilogue on GPU when available)
Copennn::ConvolutionOp	2D convolution operator (NHWC layout) backed by Eigen on CPU and cuDNN on GPU
Copennn::ConvolutionReluOp	Fused 2D convolution + ReLU activation (uses cuDNN fused epilogue on GPU)
Copennn::DropoutOp	Inverted dropout: at training time zeros activations with probability rate and rescales survivors
Copennn::EmbeddingLookupOp	Token embedding lookup with optional scaling and additive positional encoding
Copennn::FlatOp	Flattens a multi-dimensional tensor into a 2D (batch, features) tensor
Copennn::LayerNormOp	Layer normalization with learnable scale/shift, applied across the embedding dimension
Copennn::MergeOp	Reshapes (batch, heads, seq, head_dim) tensors back into (batch, seq, embed); no parameters
Copennn::MultiHeadProjectionOp	Projects (input_features) into (heads * head_dim) and reshapes for multi-head attention
Copennn::Pool3dOp	Sequence-wide 1D pooling over the embedding dimension (mean or max)
Copennn::PoolOp	2D pooling operator supporting max and average reductions
Copennn::ScaleOp	Scales inputs to a target range using per-feature minimum/maximum or mean/std statistics
Copennn::UnscaleOp	Inverse of ScaleOp: maps normalized outputs back to the original feature range
►Copennn::Optimizer	Abstract base class for training optimizers (Adam, SGD, Quasi-Newton, Levenberg-Marquardt)
Copennn::AdaptiveMomentEstimation	Adam optimizer with first/second gradient moments for stochastic minibatch training
Copennn::LevenbergMarquardtAlgorithm	Levenberg-Marquardt optimizer combining Gauss-Newton and gradient descent through an adaptive damping factor
Copennn::QuasiNewtonMethod	Quasi-Newton optimizer using the BFGS inverse-Hessian approximation with line search
Copennn::StochasticGradientDescent	Stochastic gradient descent with optional momentum, Nesterov, and learning-rate decay
Copennn::OptimizerData	Per-optimizer scratch state (moments, directions, iteration counter) backing the update step
Copennn::Registry< T >
Copennn::Configuration::Resolved	Resolved configuration after Auto values are mapped to concrete device and types
Copennn::ResponseOptimization	Optimizes input values so that a network's outputs satisfy user-defined conditions and objectives
Copennn::CsvReader::Result	Parsed CSV result; owns the source buffer that backs all row views
Copennn::TestingAnalysis::RocAnalysis	Results of a ROC analysis: ROC curve, area under it and optimal threshold
Copennn::TransformerDecoder::SamplingConfig	Sampling parameters that control how the next token is drawn from the model output distribution
Copennn::ScopedTimer
Copennn::Shape	Fixed-capacity small-vector describing tensor dimensions (rank up to MaxRank)
Copennn::Stats
Copennn::TensorSpec	Lightweight description of a tensor's shape and data type (no storage attached)
Copennn::TensorView	Non-owning view over a tensor: pointer, shape, and data type with rich reshape helpers
Copennn::TestingAnalysis	Performs post-training analysis of a neural network: errors, confusion matrices, ROC, gain charts, etc
Copennn::ThreadSafeQueue< T >
Copennn::TrainingResults	History and final metrics produced by a training run
Copennn::TrainingStrategy	High-level orchestrator pairing a Loss with an Optimizer for a network/dataset
Copennn::TransformerDecoder	Drives token-by-token inference of a Transformer model with configurable sampling strategies
Copennn::TypeInfo< T >	Compile-time traits mapping an opennn::Type to its underlying numeric type and library identifiers
Copennn::TypeInfo< Type::BF16 >	TypeInfo specialization for bfloat16 (BF16) tensors
Copennn::TypeInfo< Type::FP32 >	TypeInfo specialization for 32-bit floating point (FP32) tensors
Copennn::TypeInfo< Type::INT8 >	TypeInfo specialization for signed 8-bit integer (INT8) tensors
Copennn::Variable	Single dataset column descriptor: name, role, type, scaler, and optional categories