MNPE class similar to MNLE #1362
Conversation
|
Update:
|
…ODO: MNPE class + test
…dent; embedding net in mnpe not working yet
…not allow gpu handling yet though
|
Updates:
|
| This estimator combines a Categorical net and a neural density estimator to model | ||
| data with mixed types (discrete and continuous), e.g., as they occur in | ||
| decision-making models. It can be used for both likelihood and posterior estimation | ||
| of mixed data. |
There was a problem hiding this comment.
| This estimator combines a Categorical net and a neural density estimator to model | |
| data with mixed types (discrete and continuous), e.g., as they occur in | |
| decision-making models. It can be used for both likelihood and posterior estimation | |
| of mixed data. | |
| This estimator combines a categorical mass estimator and a density estimator to model | |
| variables with mixed types (discrete and continuous). It can be used for both likelihood | |
| estimation (e.g., for discrete decisions and continuous reaction times in decision-making | |
| models) or posterior estimation (e.g., for models that have both discrete and continuous | |
| parameters). |
| """The forward method is not implemented for MNLE, use '.sample(...)' to | ||
| generate samples though a forward pass.""" | ||
| """The forward method is not implemented for mixed neural density | ||
| estimation,use '.sample(...)' to generate samples though a forward |
There was a problem hiding this comment.
| estimation,use '.sample(...)' to generate samples though a forward | |
| estimation, use '.sample(...)' to generate samples though a forward |
| def build_mnle( | ||
| batch_x: Tensor, | ||
| batch_y: Tensor, | ||
| **kwargs, | ||
| ) -> MixedDensityEstimator: | ||
| """Returns a mixed neural likelihood estimator. | ||
|
|
||
| This estimator models p(x|theta) where x contains both continuous and discrete data. | ||
|
|
||
| Args: | ||
| batch_x: Batch of xs (data), used to infer dimensionality. | ||
| batch_y: Batch of ys (parameters), used to infer dimensionality. | ||
| **kwargs: Additional arguments passed to _build_mixed_density_estimator. | ||
|
|
||
| Returns: | ||
| MixedDensityEstimator for MNLE. | ||
| """ | ||
| return _build_mixed_density_estimator( | ||
| batch_x=batch_x, batch_y=batch_y, mode="mnle", **kwargs | ||
| ) | ||
|
|
||
|
|
||
| def build_mnpe( | ||
| batch_x: Tensor, | ||
| batch_y: Tensor, | ||
| **kwargs, | ||
| ) -> MixedDensityEstimator: | ||
| """Returns a mixed neural posterior estimator. | ||
|
|
||
| This estimator models p(theta|x) where x contains both continuous and discrete data. |
There was a problem hiding this comment.
I am confused by these two call functions. maybe I am missing something, but couldn't we just call _build_mixed_density_estimator with batch_x and batch_y swapped for MNPE and MNLE?
To me it seems this swapping is not happening, i.e., we need to make sure that in MNPE we are only embedding x and not theta, and v.v. in MNLE. Let's discuss tomorrow.
There was a problem hiding this comment.
True, we could remove both functions and just use _build_mixed_density_estimator.
The swapping of x/theta is happening because the function is once called as likelihood_nn and once as posterior_nn.
|
Update:
|
sbi/utils/sbiutils.py
Outdated
| try: | ||
| prior_mean = prior.mean.to(device) | ||
| prior_std = prior.stddev.to(device) | ||
| except (NotImplementedError, AttributeError): | ||
| warnings.warn( | ||
| "The passed discrete prior has no mean or stddev attribute, " | ||
| "estimating them from samples to build affine standardizing " | ||
| "transform.", | ||
| stacklevel=2, | ||
| ) | ||
| theta = prior.sample(torch.Size((num_prior_samples_for_zscoring,))) | ||
| prior_mean = theta.mean(dim=0).to(device) | ||
| prior_std = theta.std(dim=0).to(device) |
There was a problem hiding this comment.
move this into a small function to avoid code duplication?
tests/mnpe_test.py
Outdated
| x = mixed_param_simulator(theta) | ||
|
|
||
| # Build estimator manually | ||
| theta_embedding = FCEmbedding(1, 1) # simple embedding net, 1 continuous parameter |
There was a problem hiding this comment.
this should be an x_embedding to avoid confusion
tests/mnpe_test.py
Outdated
| log_transform_x=False, | ||
| ) | ||
| trainer = MNPE(density_estimator=density_estimator) | ||
| trainer.append_simulations(theta, x).train(max_num_epochs=5) |
There was a problem hiding this comment.
max_num_epochs=1
to speed up tests.
There was a problem hiding this comment.
please remove all diffs here. we probably want to have some kind of tutorial or how-to-guide with MNPE, but let's wait for the new documentation setup.
|
Update: |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1362 +/- ##
===========================================
- Coverage 89.62% 79.09% -10.53%
===========================================
Files 121 122 +1
Lines 9347 9394 +47
===========================================
- Hits 8377 7430 -947
- Misses 970 1964 +994
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
tests/mnpe_test.py
Outdated
| theta_true = torch.cat( | ||
| ( | ||
| torch.rand(batch_size, 1), | ||
| torch.ones(batch_size, 1), | ||
| torch.rand(batch_size, 2), | ||
| torch.bernoulli(0.8 * torch.ones(batch_size, 1)), | ||
| ), | ||
| dim=1, | ||
| ) |
There was a problem hiding this comment.
you could just do prior.sample((1,)) no?
There was a problem hiding this comment.
we usually call it theta_o :)
Implementation of mixed NPE where we have some continuous parameters theta followed by one (or multiple with this PR #1269) discrete parameters. The observation space is fully continuous.
Deprecated mnle.py in net_builders und unified mnle/mnpe as mixed_nets.py.