simple_scvi.MyModel#
- class simple_scvi.MyModel(adata, n_hidden=128, n_latent=10, n_layers=1, **model_kwargs)#
Skeleton for an scvi-tools model.
Please use this skeleton to create new models. This is a simple implementation of the scVI model [LRC+18].
- Parameters:
adata (
AnnData
) – AnnData object that has been registered viasetup_anndata()
.n_hidden (
int
(default:128
)) – Number of nodes per hidden layer.n_latent (
int
(default:10
)) – Dimensionality of the latent space.n_layers (
int
(default:1
)) – Number of hidden layers used for encoder and decoder NNs.**model_kwargs – Keyword args for
MyModule
Examples
>>> adata = anndata.read_h5ad(path_to_anndata) >>> mypackage.MyModel.setup_anndata(adata, batch_key="batch") >>> vae = mypackage.MyModel(adata) >>> vae.train() >>> adata.obsm["X_mymodel"] = vae.get_latent_representation()
Attributes table#
Data attached to model instance. |
|
Manager instance associated with self.adata. |
|
The current device that the module's params are on. |
|
Returns computed metrics during training. |
|
Whether the model has been trained. |
|
Summary string of the model. |
|
Observations that are in test set. |
|
Observations that are in train set. |
|
Observations that are in validation set. |
Methods table#
|
Converts a legacy saved model (<v0.15.0) to the updated save format. |
|
Deregisters the |
|
Retrieves the |
|
Return the ELBO for the data. |
|
Returns the object in AnnData associated with the key in the data registry. |
|
Return the latent representation for each cell. |
|
Return the marginal LL for the data. |
|
Return the reconstruction error for the data. |
|
Instantiate a model from the saved output. |
|
Return the full registry saved with the model. |
|
Registers an |
|
Save the state of the model. |
|
Sets up the |
|
Move model to device. |
|
Train the model. |
|
Print summary of the setup for the initial AnnData or a given AnnData object. |
|
Print args used to setup a saved model. |
Attributes#
adata#
- MyModel.adata#
Data attached to model instance.
adata_manager#
- MyModel.adata_manager#
Manager instance associated with self.adata.
device#
- MyModel.device#
The current device that the module’s params are on.
history#
- MyModel.history#
Returns computed metrics during training.
is_trained#
- MyModel.is_trained#
Whether the model has been trained.
summary_string#
- MyModel.summary_string#
Summary string of the model.
test_indices#
- MyModel.test_indices#
Observations that are in test set.
train_indices#
- MyModel.train_indices#
Observations that are in train set.
validation_indices#
- MyModel.validation_indices#
Observations that are in validation set.
Methods#
convert_legacy_save#
- classmethod MyModel.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 directory where legacy model is saved.output_dir_path (
str
) – Path to save converted save files.overwrite (
bool
(default:False
)) – Overwrite existing data or not. IfFalse
and directory already exists atoutput_dir_path
, error will be raised.prefix (
Optional
[str
] (default:None
)) – Prefix of saved file names.**save_kwargs – Keyword arguments passed into
save()
.
- Return type:
None
deregister_manager#
- MyModel.deregister_manager(adata=None)#
Deregisters the
AnnDataManager
instance associated withadata
.If
adata
isNone
, deregisters allAnnDataManager
instances in both the class and instance-specific manager stores, except for the one associated with this model instance.
get_anndata_manager#
- MyModel.get_anndata_manager(adata, required=False)#
Retrieves the
AnnDataManager
for a given AnnData object specific to this model instance.Requires
self.id
has been set. Checks for anAnnDataManager
specific to this model instance.- Parameters:
adata (
Union
[AnnData
,MuData
]) – AnnData object to find manager instance for.required (
bool
(default:False
)) – If True, errors on missing manager. Otherwise, returns None when manager is missing.
- Return type:
AnnDataManager
|None
get_elbo#
- MyModel.get_elbo(adata=None, indices=None, batch_size=None)#
Return the ELBO for the data.
The ELBO is a lower bound on the log likelihood of the data used for optimization of VAEs. Note, this is not the negative ELBO, higher is better.
- Parameters:
adata (
Optional
[AnnData
] (default:None
)) – AnnData object with equivalent structure to initial AnnData. IfNone
, defaults to the AnnData object used to initialize the model.indices (
Optional
[Sequence
[int
]] (default:None
)) – Indices of cells in adata to use. IfNone
, all cells are used.batch_size (
Optional
[int
] (default:None
)) – Minibatch size for data loading into model. Defaults toscvi.settings.batch_size
.
- Return type:
float
get_from_registry#
- MyModel.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.
get_latent_representation#
- MyModel.get_latent_representation(adata=None, indices=None, give_mean=True, mc_samples=5000, batch_size=None, return_dist=False)#
Return the latent representation for each cell.
This is typically denoted as \(z_n\).
- Parameters:
adata (
Optional
[AnnData
] (default:None
)) – AnnData object with equivalent structure to initial AnnData. IfNone
, defaults to the AnnData object used to initialize the model.indices (
Optional
[Sequence
[int
]] (default:None
)) – Indices of cells in adata to use. IfNone
, all cells are used.give_mean (
bool
(default:True
)) – Give mean of distribution or sample from it.mc_samples (
int
(default:5000
)) – For distributions with no closed-form mean (e.g.,logistic normal
), how many Monte Carlo samples to take for computing mean.batch_size (
Optional
[int
] (default:None
)) – Minibatch size for data loading into model. Defaults toscvi.settings.batch_size
.return_dist (
bool
(default:False
)) – Return (mean, variance) of distributions instead of just the mean. IfTrue
, ignoresgive_mean
andmc_samples
. In the case of the latter,mc_samples
is used to compute the mean of a transformed distribution. Ifreturn_dist
is true the untransformed mean and variance are returned.
- Return type:
- Returns:
: Low-dimensional representation for each cell or a tuple containing its mean and variance.
get_marginal_ll#
- MyModel.get_marginal_ll(adata=None, indices=None, n_mc_samples=1000, batch_size=None, return_mean=True, **kwargs)#
Return the marginal LL for the data.
The computation here is a biased estimator of the marginal log likelihood of the data. Note, this is not the negative log likelihood, higher is better.
- Parameters:
adata (
Optional
[AnnData
] (default:None
)) – AnnData object with equivalent structure to initial AnnData. IfNone
, defaults to the AnnData object used to initialize the model.indices (
Optional
[Sequence
[int
]] (default:None
)) – Indices of cells in adata to use. IfNone
, all cells are used.n_mc_samples (
int
(default:1000
)) – Number of Monte Carlo samples to use for marginal LL estimation.batch_size (
Optional
[int
] (default:None
)) – Minibatch size for data loading into model. Defaults toscvi.settings.batch_size
.return_mean (
Optional
[bool
] (default:True
)) – If False, return the marginal log likelihood for each observation. Otherwise, return the mmean arginal log likelihood.
- Return type:
Union
[Tensor
,float
]
get_reconstruction_error#
- MyModel.get_reconstruction_error(adata=None, indices=None, batch_size=None)#
Return the reconstruction error for the data.
This is typically written as \(p(x \mid z)\), the likelihood term given one posterior sample. Note, this is not the negative likelihood, higher is better.
- Parameters:
adata (
Optional
[AnnData
] (default:None
)) – AnnData object with equivalent structure to initial AnnData. IfNone
, defaults to the AnnData object used to initialize the model.indices (
Optional
[Sequence
[int
]] (default:None
)) – Indices of cells in adata to use. IfNone
, all cells are used.batch_size (
Optional
[int
] (default:None
)) – Minibatch size for data loading into model. Defaults toscvi.settings.batch_size
.
- Return type:
float
load#
- classmethod MyModel.load(dir_path, adata=None, accelerator='auto', device='auto', prefix=None, backup_url=None)#
Instantiate a model from the saved output.
- Parameters:
dir_path (
str
) – Path to saved outputs.adata (
Union
[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 savedscvi
setup dictionary. If None, will check for and load anndata saved with the model.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
orstr
) or"auto"
for automatic selection based on the chosen accelerator. If set to"auto"
andaccelerator
is not determined to be"cpu"
, thendevice
will be set to the first available device.prefix (
Optional
[str
] (default:None
)) – Prefix of saved file names.backup_url (
Optional
[str
] (default:None
)) – URL to retrieve saved outputs from if not present on disk.
- Returns:
: Model with loaded state dictionaries.
Examples
>>> model = ModelClass.load(save_path, adata) # use the name of the model class used to save >>> model.get_....
load_registry#
- static MyModel.load_registry(dir_path, prefix=None)#
Return the full registry saved with the model.
- Parameters:
dir_path (
str
) – Path to saved outputs.prefix (
Optional
[str
] (default:None
)) – Prefix of saved file names.
- Return type:
dict
- Returns:
: The full registry saved with the model
register_manager#
- classmethod MyModel.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 thesetup_anndata()
class method followed up by retrieval of theAnnDataManager
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 previousAnnDataManager
.
save#
- MyModel.save(dir_path, prefix=None, overwrite=False, save_anndata=False, save_kwargs=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 (
Optional
[str
] (default:None
)) – Prefix to prepend to saved file names.overwrite (
bool
(default:False
)) – Overwrite existing data or not. IfFalse
and directory already exists atdir_path
, error will be raised.save_anndata (
bool
(default:False
)) – If True, also saves the anndatasave_kwargs (
Optional
[dict
] (default:None
)) – Keyword arguments passed intosave()
.anndata_write_kwargs – Kwargs for
write()
setup_anndata#
- classmethod MyModel.setup_anndata(adata, batch_key=None, labels_key=None, layer=None, categorical_covariate_keys=None, continuous_covariate_keys=None, **kwargs)#
Sets up the
AnnData
object for this model.A mapping will be created between data fields used by this model to their respective locations in adata. None of the data in adata are modified. Only adds fields to adata.
- Parameters:
adata (
AnnData
) – AnnData object. Rows represent cells, columns represent features.batch_key (
Optional
[str
] (default:None
)) – key inadata.obs
for batch information. Categories will automatically be converted into integer categories and saved toadata.obs['_scvi_batch']
. IfNone
, assigns the same batch to all the data.labels_key (
Optional
[str
] (default:None
)) – key inadata.obs
for label information. Categories will automatically be converted into integer categories and saved toadata.obs['_scvi_labels']
. IfNone
, assigns the same label to all the data.layer (
Optional
[str
] (default:None
)) – if notNone
, uses this as the key inadata.layers
for raw count data.categorical_covariate_keys (
Optional
[List
[str
]] (default:None
)) – keys inadata.obs
that correspond to categorical data. These covariates can be added in addition to the batch covariate and are also treated as nuisance factors (i.e., the model tries to minimize their effects on the latent space). Thus, these should not be used for biologically-relevant factors that you do _not_ want to correct for.continuous_covariate_keys (
Optional
[List
[str
]] (default:None
)) – keys inadata.obs
that correspond to continuous data. These covariates can be added in addition to the batch covariate and are also treated as nuisance factors (i.e., the model tries to minimize their effects on the latent space). Thus, these should not be used for biologically-relevant factors that you do _not_ want to correct for.
- Return type:
Optional
[AnnData
]- Returns:
: None. Adds the following fields:
- .uns[‘_scvi’]
scvi
setup dictionary- .obs[‘_scvi_labels’]
labels encoded as integers
- .obs[‘_scvi_batch’]
batch encoded as integers
to_device#
- MyModel.to_device(device)#
Move model to device.
- Parameters:
device (
str
|int
) – Device to move model to. Options: ‘cpu’ for CPU, integer GPU index (eg. 0), or ‘cuda:X’ where X is the GPU index (eg. ‘cuda:0’). See torch.device for more info.
Examples
>>> adata = scvi.data.synthetic_iid() >>> model = scvi.model.SCVI(adata) >>> model.to_device('cpu') # moves model to CPU >>> model.to_device('cuda:0') # moves model to GPU 0 >>> model.to_device(0) # also moves model to GPU 0
train#
- MyModel.train(max_epochs=None, accelerator='auto', devices='auto', train_size=0.9, validation_size=None, shuffle_set_split=True, load_sparse_tensor=False, batch_size=128, early_stopping=False, datasplitter_kwargs=None, plan_kwargs=None, data_module=None, **trainer_kwargs)#
Train the model.
- Parameters:
max_epochs (
Optional
[int
] (default:None
)) – The maximum number of epochs to train the model. The actual number of epochs may be less if early stopping is enabled. IfNone
, defaults to a heuristic based onget_max_epochs_heuristic()
. Must be passed in ifdata_module
is passed in, and it does not have ann_obs
attribute.accelerator (
str
(default:'auto'
)) – Supports passing different accelerator types("cpu", "gpu", "tpu", "ipu", "hpu", "mps, "auto")
as well as custom accelerator instances.devices (
int
|list
[int
] |str
(default:'auto'
)) – The devices to use. Can be set to a non-negative index (int
orstr
), a sequence of device indices (list
or comma-separatedstr
), the value-1
to indicate all available devices, or"auto"
for automatic selection based on the chosenaccelerator
. If set to"auto"
andaccelerator
is not determined to be"cpu"
, thendevices
will be set to the first available device.train_size (
float
(default:0.9
)) – Size of training set in the range[0.0, 1.0]
. Passed intoDataSplitter
. Not used ifdata_module
is passed in.validation_size (
Optional
[float
] (default:None
)) – Size of the test set. IfNone
, defaults to1 - train_size
. Iftrain_size + validation_size < 1
, the remaining cells belong to a test set. Passed intoDataSplitter
. Not used ifdata_module
is passed in.shuffle_set_split (
bool
(default:True
)) – Whether to shuffle indices before splitting. IfFalse
, the val, train, and test set are split in the sequential order of the data according tovalidation_size
andtrain_size
percentages. Passed intoDataSplitter
. Not used ifdata_module
is passed in.load_sparse_tensor (
bool
(default:False
)) –EXPERIMENTAL
IfTrue
, loads data with sparse CSR or CSC layout as aTensor
with the same layout. Can lead to speedups in data transfers to GPUs, depending on the sparsity of the data. Passed intoDataSplitter
. Not used ifdata_module
is passed in.batch_size (
Tunable_
[int
] (default:128
)) – Minibatch size to use during training. Passed intoDataSplitter
. Not used ifdata_module
is passed in.early_stopping (
bool
(default:False
)) – Perform early stopping. Additional arguments can be passed in through**kwargs
. SeeTrainer
for further options.datasplitter_kwargs (
Optional
[dict
] (default:None
)) – Additional keyword arguments passed intoDataSplitter
. Values in this argument can be overwritten by arguments directly passed into this method, when appropriate. Not used ifdata_module
is passed in.plan_kwargs (
Optional
[dict
] (default:None
)) – Additional keyword arguments passed intoTrainingPlan
. Values in this argument can be overwritten by arguments directly passed into this method, when appropriate.data_module (
Optional
[LightningDataModule
] (default:None
)) –EXPERIMENTAL
ALightningDataModule
instance to use for training in place of the defaultDataSplitter
. Can only be passed in if the model was not initialized withAnnData
.**kwargs – Additional keyword arguments passed into
Trainer
.
view_anndata_setup#
- MyModel.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 (
Union
[AnnData
,MuData
,None
] (default:None
)) – AnnData object setup withsetup_anndata
ortransfer_fields()
.hide_state_registries (
bool
(default:False
)) – If True, prints a shortened summary without details of each state registry.
- Return type:
None
view_setup_args#
- static MyModel.view_setup_args(dir_path, prefix=None)#
Print args used to setup a saved model.
- Parameters:
dir_path (
str
) – Path to saved outputs.prefix (
Optional
[str
] (default:None
)) – Prefix of saved file names.
- Return type:
None