decima.hub package

Submodules

decima.hub.download module

decima.hub.download.download_decima_data()

Download all required data for Decima.

decima.hub.download.download_decima_metadata()

Download pre-trained Decima model data from wandb.

decima.hub.download.download_decima_weights()

Download pre-trained Decima model weights from wandb.

decima.hub.download.download_hg38()

Download hg38 genome from UCSC.

Module contents

decima.hub.load_decima_metadata(path=None)

Load the Decima metadata from wandb.

Parameters:

path (Optional[str]) – Path to local metadata file. If None, downloads from wandb.

Returns:

An AnnData object containing the Decima metadata.

decima.hub.load_decima_model(model=0, device=None)

Load a pre-trained Decima model from wandb or local path.

Parameters:

• model (Union[str, int]) – Model identifier or path. Can be: - int: Replicate number (0-3) - str: Model name on wandb - str: Path to local model checkpoint

• device (Optional[str]) – Device to load the model on. If None, automatically selects the best available device.

Returns:

A pre-trained Decima model instance loaded on the specified device.

Return type:

LightningModel

Raises:

ValueError – If model identifier is invalid or not found.

decima.hub.login_wandb()