scvi.distributions.Log1pNormal

scvi.distributions.Log1pNormal#

class scvi.distributions.Log1pNormal(mu, sigma, scale=None, validate_args=False)[source]#

Log1p-Normal distribution.

Distribution where log(X + 1) follows a Normal distribution. This allows modeling data with zeros, as the support is [0, ∞).

In the (mu, sigma) parameterization, samples from the log1p-normal are generated as follows:

\(z \sim \textrm{Normal}(\mu, \sigma)\)
\(x = \exp(z) - 1\)

The probability density function is:

\[f(x; \mu, \sigma) = \frac{1}{(x+1) \sigma \sqrt{2\pi}} \exp\left(-\frac{(\ln(x+1) - \mu)^2}{2\sigma^2}\right)\]

Parameters:

mu (Tensor) – Mean of the normal distribution in log1p space.
sigma (Tensor) – Standard deviation of the normal distribution in log1p space.
scale (Tensor | None (default: None)) – Normalized mean expression of the distribution. This optional parameter is not used in any computations but allows storing normalization expression levels.
validate_args (bool (default: False)) – Raise ValueError if arguments do not match constraints.

Attributes table#

`arg_constraints`
`batch_shape`	Returns the shape over which parameters are batched.
`event_shape`	Returns the shape of a single sample (without batching).
`has_enumerate_support`
`has_rsample`
`mean`	Mean of the distribution.
`mode`	Returns the mode of the distribution.
`stddev`	Returns the standard deviation of the distribution.
`support`
`variance`	Variance of the distribution.

Methods table#

`cdf`(value)	Returns the cumulative density/mass function evaluated at value.
`entropy`()	Returns entropy of distribution, batched over batch_shape.
`enumerate_support`([expand])	Returns tensor containing all values supported by a discrete distribution.
`expand`(batch_shape[, _instance])	Returns a new distribution instance (or populates an existing instance provided by a derived class) with batch dimensions expanded to batch_shape.
`get_normalized`(key)	Get normalized values.
`icdf`(value)	Returns the inverse cumulative density/mass function evaluated at value.
`log_prob`(value)	Log probability with log1p transformation.
`perplexity`()	Returns perplexity of distribution, batched over batch_shape.
`rsample`([sample_shape])	Generates a sample_shape shaped reparameterized sample or sample_shape shaped batch of reparameterized samples if the distribution parameters are batched.
`sample`([sample_shape])	Sample from the distribution.
`sample_n`(n)	Generates n samples or n batches of samples if the distribution parameters are batched.
`set_default_validate_args`(value)	Sets whether validation is enabled or disabled.

Attributes#

Log1pNormal.arg_constraints = {'mu': Optional(Real()), 'scale': Optional(GreaterThanEq(lower_bound=0)), 'sigma': Optional(GreaterThan(lower_bound=0))}#

Log1pNormal.batch_shape[source]#: Returns the shape over which parameters are batched.

Log1pNormal.event_shape[source]#: Returns the shape of a single sample (without batching).

Log1pNormal.has_enumerate_support = False#

Log1pNormal.has_rsample = False#

Log1pNormal.mean[source]#: Mean of the distribution.

Log1pNormal.mode[source]#: Returns the mode of the distribution.

Log1pNormal.stddev[source]#: Returns the standard deviation of the distribution.

Log1pNormal.support = GreaterThanEq(lower_bound=0.0)#

Log1pNormal.variance[source]#: Variance of the distribution.

Methods#

Log1pNormal.cdf(value)[source]#

Returns the cumulative density/mass function evaluated at value.

Parameters:: value (Tensor)
Return type:: Tensor

Log1pNormal.entropy()[source]#

Returns entropy of distribution, batched over batch_shape.

Return type:: Tensor
Returns:: Tensor of shape batch_shape.

Log1pNormal.enumerate_support(expand=True)[source]#

Returns tensor containing all values supported by a discrete distribution. The result will enumerate over dimension 0, so the shape of the result will be (cardinality,) + batch_shape + event_shape (where event_shape = () for univariate distributions).

Note that this enumerates over all batched tensors in lock-step [[0, 0], [1, 1], …]. With expand=False, enumeration happens along dim 0, but with the remaining batch dimensions being singleton dimensions, [[0], [1], ...

To iterate over the full Cartesian product use itertools.product(m.enumerate_support()).

Parameters:: expand (bool) – whether to expand the support over the batch dims to match the distribution’s batch_shape.
Return type:: Tensor
Returns:: Tensor iterating over dimension 0.

Log1pNormal.expand(batch_shape, _instance=None)[source]#

Returns a new distribution instance (or populates an existing instance provided by a derived class) with batch dimensions expanded to batch_shape. This method calls expand on the distribution’s parameters. As such, this does not allocate new memory for the expanded distribution instance. Additionally, this does not repeat any args checking or parameter broadcasting in __init__.py, when an instance is first created.

Parameters:

batch_shape (torch.Size) – the desired expanded size.
_instance (default: None) – new instance provided by subclasses that need to override .expand.

Returns:

New distribution instance with batch dimensions expanded to batch_size.

Log1pNormal.get_normalized(key)[source]#

Get normalized values.

Return type:: Tensor

Log1pNormal.icdf(value)[source]#

Returns the inverse cumulative density/mass function evaluated at value.

Parameters:: value (Tensor)
Return type:: Tensor

Log1pNormal.log_prob(value)[source]#

Log probability with log1p transformation.

Return type:: Tensor

Log1pNormal.perplexity()[source]#

Returns perplexity of distribution, batched over batch_shape.

Return type:: Tensor
Returns:: Tensor of shape batch_shape.

Log1pNormal.rsample(sample_shape=())[source]#

Generates a sample_shape shaped reparameterized sample or sample_shape shaped batch of reparameterized samples if the distribution parameters are batched.

Return type:: Tensor

Log1pNormal.sample(sample_shape=None)[source]#

Sample from the distribution.

Return type:: Tensor

Log1pNormal.sample_n(n)[source]#

Generates n samples or n batches of samples if the distribution parameters are batched.

Return type:: Tensor

static Log1pNormal.set_default_validate_args(value)[source]#

Sets whether validation is enabled or disabled.

The default behavior mimics Python’s assert statement: validation is on by default, but is disabled if Python is run in optimized mode (via python -O). Validation may be expensive, so you may want to disable it once a model is working.

Parameters:: value (bool) – Whether to enable validation.
Return type:: None