Low-level Modules¶

This section documents the low-level, fine-grained modules of OmniGenBench, including models, datasets, metrics, tokenizers, and trainers. These modules provide the building blocks for custom development and advanced usage.

Overview:

Low-level modules expose the core functionality and implementation details of OmniGenBench.
Users can explore these modules to understand the internal design, extend functionality, or troubleshoot specific issues.
Each module is documented with its classes, functions, and usage examples.

Categories:

Models: Classification, regression, and specialized genomic models.
Datasets: Classes for handling various genomic data formats.
Metrics: Evaluation metrics for classification, regression, and ranking tasks.
Tokenizers: Sequence representation and preprocessing tools.
Trainers: Training utilities for different frameworks and workflows.

Refer to the documentation below for details on each module, including available classes, methods, and extension options.