core.posterior_models.score_matching.ScoreDiffusionPosteriorModel
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel(**kwargs)Attributes
| Name | Description |
|---|---|
| beta_max | |
| beta_min | |
| eps | |
| likelihood_weighting |
Methods
| Name | Description |
|---|---|
| alpha | |
| beta | |
| evaluate_vector_field | Evaluate the vector field v(t, theta_t, context_data) that generates the flow |
| get_likelihood_weighting | |
| get_t_theta_t_score | |
| loss | Returns the score matching loss for parameters theta conditioned on context. |
| mu | |
| sigma |
alpha
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.alpha(t)beta
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.beta(t)evaluate_vector_field
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.evaluate_vector_field(
t,
theta_t,
*context_data,
)Evaluate the vector field v(t, theta_t, context_data) that generates the flow via the ODE
d/dt f(theta_t, t, context) = v(f(theta_t, t, context), t, context).For score matching, the vector field (or drift function) is computed from the predicted score.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| t | time (noise level) | required | |
| theta_t | noisy parameters, perturbed with noise level t | required | |
| *context_data | list with context data (GW data) | () |
get_likelihood_weighting
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.get_likelihood_weighting(
weighting,
)get_t_theta_t_score
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.get_t_theta_t_score(
theta_1,
)loss
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.loss(
theta,
*context_data,
)Returns the score matching loss for parameters theta conditioned on context.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
| theta | parameters (e.g., binary-black hole parameters) | required | |
| *context_data | context data (e.g., gravitational-wave data) | () |
Returns
| Name | Type | Description |
|---|---|---|
| torch.tensor | Loss. |
mu
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.mu(t, x_1)sigma
core.posterior_models.score_matching.ScoreDiffusionPosteriorModel.sigma(t)