scvi.external.tangram.TangramMapper#

class scvi.external.tangram.TangramMapper(n_obs_sc, n_obs_sp, lambda_g1=1.0, lambda_d=0.0, lambda_g2=0.0, lambda_r=0.0, lambda_count=1.0, lambda_f_reg=1.0, constrained=False, target_count=None, training=True, parent=<flax.linen.module._Sentinel object>, name=None)[source]#

Bases: JaxBaseModuleClass

Tangram Mapper Model.

Parameters:

n_obs_sc (int) –
n_obs_sp (int) –
lambda_g1 (float) –
lambda_d (float) –
lambda_g2 (float) –
lambda_r (float) –
lambda_count (float) –
lambda_f_reg (float) –
constrained (bool) –
target_count (Optional[int]) –
training (bool) –
parent (Optional[Union[Type[Module], Type[Scope], Type[_Sentinel]]]) –
name (Optional[str]) –

Attributes table#

`constrained`
`lambda_count`
`lambda_d`
`lambda_f_reg`
`lambda_g1`
`lambda_g2`
`lambda_r`
`name`
`parent`
`required_rngs`	Returns a tuple of rng sequence names required for this Flax module.
`scope`
`target_count`
`training`
`n_obs_sc`
`n_obs_sp`

Methods table#

`generative`()	No generative model here.
`inference`()	Run inference model.
`loss`(tensors, inference_outputs, ...)	Compute loss.
`setup`()	Setup model.

Attributes#

constrained

TangramMapper.constrained: bool = False#

lambda_count

TangramMapper.lambda_count: float = 1.0#

lambda_d

TangramMapper.lambda_d: float = 0.0#

lambda_f_reg

TangramMapper.lambda_f_reg: float = 1.0#

lambda_g1

TangramMapper.lambda_g1: float = 1.0#

lambda_g2

TangramMapper.lambda_g2: float = 0.0#

lambda_r

TangramMapper.lambda_r: float = 0.0#

name

TangramMapper.name: Optional[str] = None#

parent

TangramMapper.parent: Optional[Union[Type[Module], Type[Scope], Type[_Sentinel]]] = None#

required_rngs

TangramMapper.required_rngs[source]#

scope

TangramMapper.scope: Optional[Scope] = None#

target_count

TangramMapper.target_count: Optional[int] = None#

training

TangramMapper.training: bool = True#

n_obs_sc

TangramMapper.n_obs_sc: int#

n_obs_sp

TangramMapper.n_obs_sp: int#

Methods#

generative

TangramMapper.generative()[source]#

No generative model here.

Return type:: dict

inference

TangramMapper.inference()[source]#

Run inference model.

Return type:: dict

loss

TangramMapper.loss(tensors, inference_outputs, generative_outputs)[source]#: Compute loss.

setup

TangramMapper.setup()[source]#: Setup model.