Documentation

Install mRNABench, run evaluations, inspect results, or add a dataset or model.

Start with a complete run, or go directly to the dataset, model, result, and API reference pages.

Install from PyPI

python -m pip install mrna-bench

Choose a workflow

01Evaluate a model

Load a model and dataset, generate embeddings, and run an evaluation.

02Add a dataset

Define the dataframe columns and metadata, then register the dataset.

03Add a model

Implement an EmbeddingModel adapter for one or more checkpoints.

Guide sections

Use the quickstart for an end-to-end example and the concepts guide for evaluation defaults. The Python API page lists current signatures and parameter names directly from the package source.

Section	Use it for
Start	Installation, first run, and the package vocabulary.
Run	Configuration, benchmark execution, result files, and analysis.
Extend	New datasets and model adapters.
Reference	Current Python signatures and parameter notes.