API reference

High-level functions

`generate`(user_input[, conformations, ...])	Generate STARLING ensembles and distance maps for one or more sequences.
`sequence_encoder`(sequence_dict[, ...])	Embed sequences with the STARLING encoder.
`load_ensemble`(filename[, ignore_structures])	Function to read in a STARLING ensemble from a file and return the STARLING ensemble object.
`set_compilation_options`([enabled])	Configure model compilation settings programmatically with full support for PyTorch compile parameters.

`structure.ensemble`	Ensemble representation and analysis helpers.
`structure.bme`	Bayesian Maximum Entropy (BME) reweighting for ensemble refinement.
`structure.bme_utils`
`structure.coordinates`

`frontend.ensemble_generation`
`frontend.starling_viz`

`inference`	Inference module for STARLING.
`inference.generation`
`inference.constraints`
`inference.model_loading`
`inference.evaluate_vae`
`inference.benchmark_mds`

`models.vae`
`models.diffusion`
`models.unet`
`models.vit`
`models.transformer`
`models.attention`
`models.blocks`
`models.normalization`
`models.ema`
`models.continuous_diffusion`
`models.vae_components`
`models.quantize`
`models.resnets_original`

`search`	Starling search module: FAISS-backed similarity search with filtering and reranking.
`search.search_engine`	FAISS Search Engine
`search.store`	SQLite Sequence Store
`search.builder`	FAISS Index Builder
`search.similarity_search`
`search.search_utils`	Search Utilities

`configs`
`utilities`