This module provides utility classes to iterate over data.
Implementation of the various computational layers.
The library supports already implemented loss functions, as well as a callback-based way to specify a custom loss.
Low-level math primitives.
Utility functions for ML-related tasks.
Internal abstraction to report progress during training.
Base class for all layers.
NeuralNet is the main abstraction of vectorflow.
Implementation of different stochastic optimizers.
Implementation of standard regularizers for Linear layer.
Internal serialization functions for automatic neural net ser/deser.
Utility functions used across the library.
Vectorflow is a lightweight neural network library for sparse data.