TopologicalVariationalEncoder
latent.modules.encoder.VariationalEncoder
Variational encoder. This model compresses input data by parameterizing a latent distribution that is regularized through a KL Divergence loss.
__init__(self, latent_dim=50, name='variational_encoder', initializer='glorot_normal', kld_weight=0.0001, prior='normal', latent_dist='independent', iaf_units=[256, 256], n_pseudoinputs=200, **kwargs)
special
Parameters:
Name | Type | Description | Default |
---|---|---|---|
latent_dim |
int |
Integer indicating the number of dimensions in the latent space. |
50 |
name |
str |
String indicating the name of the model. |
'variational_encoder' |
initializer |
Union[str, Callable] |
Initializer for the kernel weights matrix (see
|
'glorot_normal' |
kld_weight |
float |
Float indicating the weight of the KL Divergence regularization loss. |
0.0001 |
prior |
Literal['normal', 'iaf', 'vamp'] |
The choice of prior distribution. One of the following:
|
'normal' |
latent_dist |
Literal['independent', 'multivariate'] |
The choice of latent distribution. One of the following:
|
'independent' |
iaf_units |
Iterable[int] |
Integer list indicating the units in the IAF bijector network.
Only used if |
[256, 256] |
n_pseudoinputs |
int |
Integer indicating the number of pseudoinputs for the VAMP
prior. Only used if |
200 |
**kwargs |
|
Other arguments passed on to |
{} |