scvi.model.mlxSCVI

class scvi.model.mlxSCVI(adata, n_hidden=128, n_latent=10, dropout_rate=0.1, gene_likelihood='nb', **model_kwargs)

Single-cell variational inference model using the MLX framework.

This implementation leverages the features of the MLX framework to provide optimized performance on Apple Silicon chips.

Parameters:

• adata (AnnData) – AnnData object registered via mlxSCVI.setup_anndata().

• n_hidden (int (default: 128)) – Number of nodes per hidden layer.

• n_latent (int (default: 10)) – Dimensionality of the latent space.

• dropout_rate (float (default: 0.1)) – Dropout rate for neural networks.

• gene_likelihood (Literal['nb', 'poisson'] (default: 'nb')) – One of: * ‘nb’ - Negative binomial distribution * ‘poisson’ - Poisson distribution

Attributes table

adata	Data attached to model instance.
adata_manager	Manager instance associated with self.adata.
device	Get the current device.
get_normalized_function_name	What the get normalized functions name is
history	Returns computed metrics during training.
is_trained	Whether the model has been trained.
registry	Data attached to model instance.
run_id	Returns the run id of the model.
run_name	Returns the run name of the model.
summary_string	Summary string of the model.
test_indices	Observations that are in test set.
train_indices	Observations that are in train set.
validation_indices	Observations that are in validation set.

Methods table

convert_legacy_save(dir_path, output_dir_path)	Converts a legacy saved model (<v0.15.0) to the updated save format.
data_registry(registry_key)	Returns the object in AnnData associated with the key in the data registry.
deregister_manager([adata])	Deregisters the AnnDataManager instance associated with adata.
differential_abundance(*args, **kwargs)	Not implemented for this model class.
get_anndata_manager(adata[, required])	Retrieves the AnnDataManager for a given AnnData object.
get_from_registry(adata, registry_key)	Returns the object in AnnData associated with the key in the data registry.
get_latent_representation([adata, indices, ...])	Get the latent representation for each cell.
get_normalized_expression(*args, **kwargs)	Not implemented for this model class.
get_setup_arg(setup_arg)	Returns the string provided to setup of a specific setup_arg.
get_state_registry(registry_key)	Returns the state registry for the AnnDataField registered with this instance.
get_var_names([legacy_mudata_format])	Variable names of input data.
load(dir_path[, adata, accelerator, device, ...])	Instantiate a model from the saved output.
load_registry(dir_path[, prefix])	Return the full registry saved with the model.
register_manager(adata_manager)	Registers an AnnDataManager instance with this model class.
save(dir_path[, prefix, overwrite, ...])	Save the state of the model.
setup_anndata(adata[, layer, batch_key, ...])	Set up AnnData object for training.
to_device(device)	Move the model to a specific device.
train([max_epochs, accelerator, devices, ...])	Train the model.
transfer_fields(adata, **kwargs)	Transfer fields from a model to an AnnData object.
update_setup_method_args(setup_method_args)	Update setup method args.
view_anndata_setup([adata, ...])	Print summary of the setup for the initial AnnData or a given AnnData object.
view_registry([hide_state_registries])	Prints summary of the registry.
view_setup_args(dir_path[, prefix])	Print args used to setup a saved model.
view_setup_method_args()	Prints setup kwargs used to produce a given registry.

Attributes

mlxSCVI.adata

Data attached to model instance.

mlxSCVI.adata_manager

Manager instance associated with self.adata.

mlxSCVI.device

Get the current device.

MLX automatically handles device placement.

Returns:

str Device identifier.

mlxSCVI.get_normalized_function_name

What the get normalized functions name is

mlxSCVI.history

Returns computed metrics during training.

mlxSCVI.is_trained

Whether the model has been trained.

mlxSCVI.registry

Data attached to model instance.

mlxSCVI.run_id

Returns the run id of the model. Used in MLFlow

mlxSCVI.run_name

Returns the run name of the model. Used in MLFlow

mlxSCVI.summary_string

Summary string of the model.

mlxSCVI.test_indices

Observations that are in test set.

mlxSCVI.train_indices

Observations that are in train set.

mlxSCVI.validation_indices

Observations that are in validation set.

Methods

classmethod mlxSCVI.convert_legacy_save(dir_path, output_dir_path, overwrite=False, prefix=None, **save_kwargs)

Converts a legacy saved model (<v0.15.0) to the updated save format.

Parameters:

• dir_path (str) – Path to the directory where the legacy model is saved.

• output_dir_path (str) – Path to save converted save files.

• overwrite (bool (default: False)) – Overwrite existing data or not. If False and directory already exists at output_dir_path, an error will be raised.

• prefix (str | None (default: None)) – Prefix of saved file names.

• **save_kwargs – Keyword arguments passed into save().

Return type:

None

mlxSCVI.data_registry(registry_key)

Returns the object in AnnData associated with the key in the data registry.

Parameters:

registry_key (str) – key of an object to get from self.data_registry

Return type:

ndarray | DataFrame

Returns:

The requested data.

mlxSCVI.deregister_manager(adata=None)

Deregisters the AnnDataManager instance associated with adata.

If adata is None, deregisters all AnnDataManager instances in both the class and instance-specific manager stores, except for the one associated with this model instance.

mlxSCVI.differential_abundance(*args, **kwargs)

Not implemented for this model class.

Available in models that inherit from VAEMixin.

Raises:

NotImplementedError –

mlxSCVI.get_anndata_manager(adata, required=False)

Retrieves the AnnDataManager for a given AnnData object.

Requires self.id has been set. Checks for an AnnDataManager specific to this model instance.

Parameters:

• adata (AnnData | MuData) – AnnData object to find a manager instance for.

• required (bool (default: False)) – If True, errors on missing manager. Otherwise, returns None when manager is missing.

Return type:

AnnDataManager | None

mlxSCVI.get_from_registry(adata, registry_key)

Returns the object in AnnData associated with the key in the data registry.

AnnData object should be registered with the model prior to calling this function via the self._validate_anndata method.

Parameters:

• registry_key (str) – key of object to get from the data registry.

• adata (AnnData | MuData) – AnnData to pull data from.

Return type:

ndarray

Returns:

The requested data as a NumPy array.

mlxSCVI.get_latent_representation(adata=None, indices=None, give_mean=True, n_samples=1, batch_size=None)

Get the latent representation for each cell.

Parameters:

• adata (AnnData | None (default: None)) – AnnData object with the same structure as the initial AnnData object. If None, defaults to the AnnData object used when initializing the model.

• indices (Sequence[int] | None (default: None)) – Indices of cells to use from adata. If None, all cells are used.

• give_mean (bool (default: True)) – Whether to return the mean of the posterior distribution or a sample.

• n_samples (int (default: 1)) – Number of samples to use for computing the latent representation.

• batch_size (int | None (default: None)) – Minibatch size for data loading into the model.

Return type:

ndarray

Returns:

-latent_representation (ndarray) Low-dimensional representation for each cell

mlxSCVI.get_normalized_expression(*args, **kwargs)

Not implemented for this model class.

Available in RNA models that inherit from RNASeqMixin.

Raises:

NotImplementedError –

mlxSCVI.get_setup_arg(setup_arg)

Returns the string provided to setup of a specific setup_arg.

Return type:

attrdict

mlxSCVI.get_state_registry(registry_key)

Returns the state registry for the AnnDataField registered with this instance.

Return type:

attrdict

mlxSCVI.get_var_names(legacy_mudata_format=False)

Variable names of input data.

Return type:

dict

classmethod mlxSCVI.load(dir_path, adata=None, accelerator='auto', device='auto', prefix=None, backup_url=None, datamodule=None, allowed_classes_names_list=None)

Instantiate a model from the saved output.

Parameters:

• dir_path (str) – Path to saved outputs.

• adata (AnnData | MuData | None (default: None)) – AnnData organized in the same way as data used to train model. It is not necessary to run setup_anndata, as AnnData is validated against the saved scvi setup dictionary. If None, will check for and load anndata saved with the model. If False, will load the model without AnnData.

• accelerator (str (default: 'auto')) – Supports passing different accelerator types (“cpu”, “gpu”, “tpu”, “ipu”, “hpu”, “mps, “auto”) as well as custom accelerator instances.

• device (int | str (default: 'auto')) – The device to use. Can be set to a non-negative index (int or str) or “auto” for automatic selection based on the chosen accelerator. If set to “auto” and accelerator is not determined to be “cpu”, then device will be set to the first available device.

• prefix (str | None (default: None)) – Prefix of saved file names.

• backup_url (str | None (default: None)) – URL to retrieve saved outputs from if not present on disk.

• datamodule (LightningDataModule | None (default: None)) – EXPERIMENTAL A LightningDataModule instance to use for training in place of the default DataSplitter. Can only be passed in if the model was not initialized with AnnData.

• allowed_classes_names_list (list[str] | None (default: None)) – list of allowed classes names to be loaded (besides the original class name)

Returns:

Model with loaded state dictionaries.

Examples

>>> model = ModelClass.load(save_path, adata)
>>> model.get_....

static mlxSCVI.load_registry(dir_path, prefix=None)

Return the full registry saved with the model.

Parameters:

• dir_path (str) – Path to saved outputs.

• prefix (str | None (default: None)) – Prefix of saved file names.

Return type:

dict

Returns:

The full registry saved with the model

classmethod mlxSCVI.register_manager(adata_manager)

Registers an AnnDataManager instance with this model class.

Stores the AnnDataManager reference in a class-specific manager store. Intended for use in the setup_anndata() class method followed up by retrieval of the AnnDataManager via the _get_most_recent_anndata_manager() method in the model init method.

Notes

Subsequent calls to this method with an AnnDataManager instance referring to the same underlying AnnData object will overwrite the reference to previous AnnDataManager.

mlxSCVI.save(dir_path, prefix=None, overwrite=False, save_anndata=False, save_kwargs=None, legacy_mudata_format=False, datamodule=None, **anndata_write_kwargs)

Save the state of the model.

Neither the trainer optimizer state nor the trainer history are saved. Model files are not expected to be reproducibly saved and loaded across versions until we reach version 1.0.

Parameters:

• dir_path (str) – Path to a directory.

• prefix (str | None (default: None)) – Prefix to prepend to saved file names.

• overwrite (bool (default: False)) – Overwrite existing data or not. If False and directory already exists at dir_path, an error will be raised.

• save_anndata (bool (default: False)) – If True, also saves the anndata

• save_kwargs (dict | None (default: None)) – Keyword arguments passed into save().

• legacy_mudata_format (bool (default: False)) – If True, saves the model var_names in the legacy format if the model was trained with a MuData object. The legacy format is a flat array with variable names across all modalities concatenated, while the new format is a dictionary with keys corresponding to the modality names and values corresponding to the variable names for each modality.

• datamodule (LightningDataModule | None (default: None)) – EXPERIMENTAL A LightningDataModule instance to use for training in place of the default DataSplitter. Can only be passed in if the model was not initialized with AnnData.

• anndata_write_kwargs – Kwargs for write()

classmethod mlxSCVI.setup_anndata(adata, layer=None, batch_key=None, labels_key=None, categorical_covariate_keys=None, continuous_covariate_keys=None, **kwargs)

Set up AnnData object for training.

Parameters:

• adata (AnnData) – AnnData object.

• layer (str | None (default: None)) – If not None, use this layer instead of X for training.

• batch_key (str | None (default: None)) – If not None, use the obs column specified by this key as batch information.

• labels_key (str | None (default: None)) – If not None, use the obs column specified by this key as labels.

• categorical_covariate_keys (list[str] | None (default: None)) – Keys in adata.obs for additional categorical covariates.

• continuous_covariate_keys (list[str] | None (default: None)) – Keys in adata.obs for additional continuous covariates.

mlxSCVI.to_device(device)

Move the model to a specific device.

MLX automatically handles device placement, so this is a no-op.

Parameters:

device – Target device.

mlxSCVI.train(max_epochs=None, accelerator='auto', devices='auto', train_size=None, validation_size=None, shuffle_set_split=True, batch_size=128, datasplitter_kwargs=None, plan_kwargs=None, **trainer_kwargs)

Train the model.

Parameters:

• max_epochs (int | None (default: None)) – Number of passes through the dataset. If None, defaults to np.min([round((20000 / n_cells) * 400), 400])

• accelerator (str (default: 'auto')) – Accelerator type. MLX automatically selects the best available device.

• devices (int | list[int] | str (default: 'auto')) – Device selection. MLX automatically selects the best available device.

• train_size (float | None (default: None)) – Training set size in the range [0.0, 1.0].

• validation_size (float | None (default: None)) – Validation set size. If None, defaults to 1 - train_size.

• shuffle_set_split (bool (default: True)) – Whether to shuffle indices before splitting.

• batch_size (int (default: 128)) – Minibatch size to use during training.

• datasplitter_kwargs (dict | None (default: None)) – Additional keyword arguments for DataSplitter.

• plan_kwargs (dict | None (default: None)) – Keyword arguments for the training plan.

• **trainer_kwargs – Additional keyword arguments for training.

Returns:

self The trained model instance.

mlxSCVI.transfer_fields(adata, **kwargs)

Transfer fields from a model to an AnnData object.

Return type:

AnnData

mlxSCVI.update_setup_method_args(setup_method_args)

Update setup method args.

Parameters:

setup_method_args (dict) – This is a bit of a misnomer, this is a dict representing kwargs of the setup method that will be used to update the existing values in the registry of this instance.

mlxSCVI.view_anndata_setup(adata=None, hide_state_registries=False)

Print summary of the setup for the initial AnnData or a given AnnData object.

Parameters:

• adata (AnnData | MuData | None (default: None)) – AnnData object setup with setup_anndata or transfer_fields().

• hide_state_registries (bool (default: False)) – If True, prints a shortened summary without details of each state registry.

Return type:

None

mlxSCVI.view_registry(hide_state_registries=False)

Prints summary of the registry.

Parameters:

hide_state_registries (bool (default: False)) – If True, prints a shortened summary without details of each state registry.

Return type:

None

static mlxSCVI.view_setup_args(dir_path, prefix=None)

Print args used to setup a saved model.

Parameters:

• dir_path (str) – Path to saved outputs.

• prefix (str | None (default: None)) – Prefix of saved file names.

Return type:

None

mlxSCVI.view_setup_method_args()

Prints setup kwargs used to produce a given registry.

Parameters:

registry – Registry produced by an AnnDataManager.

Return type:

None