Expand description
Neural network building blocks module for SciRS2
This module provides neural network building blocks for SciRS2, including:
- Layers (dense, convolutional, recurrent, etc.)
- Activation functions (ReLU, sigmoid, tanh, etc.)
- Loss functions (MSE, cross-entropy, etc.)
- Optimizers (SGD, Adam, etc.)
- Model architectures and training utilities
- Neural network specific linear algebra operations
- Model evaluation and testing
- Advanced training techniques
Re-exports§
pub use error::Error;pub use error::NeuralError;pub use error::Result;
Modules§
- activations
- Activation functions for neural networks
- autograd
- Automatic differentiation module for neural networks.
- callbacks
- Callback system for neural network training
- data
- Data loading and processing utilities for neural networks
- error
- Error types for the neural network module
- evaluation
- Model evaluation framework
- layers
- Neural network layers implementation
- linalg
- Neural network specific linear algebra operations
- losses
- Loss functions for neural networks
- models
- Neural network model implementations
- optimizers
- Neural network optimizers
- prelude
- Common neural network functionality
- serialization
- Module for model serialization and deserialization
- training
- Training utilities
- transformer
- Transformer models implementation
- utils
- Utility functions for neural networks