API reference

The CIFAR-10 Tools package provides a comprehensive set of modules for building, training, and evaluating convolutional neural networks on the CIFAR-10 dataset.

Core modules

• Data loading
  • Functions
  • Overview
  • Example usage
• Training
  • train_one_epoch()
  • evaluate()
  • train_model()
  • Functions
  • Overview
  • Example usage
• Evaluation
  • evaluate_model()
  • Functions
  • Overview
  • Example usage
• Plotting
  • Overview
  • Example usage
• Hyperparameter optimization
  • Overview
  • Key components
  • Example usage
  • Search space format

Module overview

Data loading (image_classification_tools.pytorch.data)

Functions for loading and preprocessing CIFAR-10 data with flexible transforms and batching.

Training (image_classification_tools.pytorch.training)

Utilities for model training with progress tracking and history logging.

Evaluation (image_classification_tools.pytorch.evaluation)

Functions for evaluating model performance and generating predictions.

Plotting (image_classification_tools.pytorch.plotting)

Visualization utilities for training curves, confusion matrices, and performance analysis.

Hyperparameter optimization (image_classification_tools.pytorch.hyperparameter_optimization)

Optuna-based hyperparameter search with configurable architectures and search spaces.

Complete module index

• Index

• Module Index