Skip to main content
Back to top
Ctrl
+
K
Getting started
Installation
Tutorials
Getting started with
sbi
The Bayesian workflow in
sbi
API of implemented methods
Examples
Inference on Hodgkin-Huxley model: tutorial
SBI for decision-making models
How-to guide
Prior and simulator
How to specify a custom prior (e.g., multiple independent priors)
How to choose summary statistics
Neural nets
How to choose neural nets
How to use embedding nets for high-dimensional observations
How to use time-series embeddings
How to use permutation-invariant embeddings for iid observations (for NPE)
How to define custom neural nets
Training
How to choose abstraction levels in sbi
How to choose an inference method
How to run sequential methods
How to use GPUs
How to save and load objects
How to resume training
How to tune hyperparameters with Optuna
How to track experiments
How to use FMPE and NPSE
Sampling
How to choose sampling algorithms
How to refine posterior estimates with importance sampling
How to sample the posterior given iid observations (for NLE and NRE)
How to visualize MCMC diagnostics with ArviZ
How to use Pyro with SBI
How to use
PosteriorParameters
in sbi
How to use FMPE and NPSE
Diagnostics
How to choose a diagnostic tool
How to run expected coverage
How to run simulation-based calibration (SBC)
How to run L-C2ST
How to run TARP
How to detect model misspecification
Visualization
How to visualize the conditional posterior distribution
Advanced tutorials
Prior and simulator
Efficient handling of invalid simulation outputs
Neural nets
Embedding nets for observations
Customizing the density estimator
SBI with iid data and permutation-invariant embeddings
Vector Field Methods: FMPE and NPSE
Training
More flexibility over the training loop and samplers
Multi-round inference
Sampling
Refining posterior estimates with importance sampling
Diagnostics
Posterior Predictive Checks (PPC) in SBI
Simulation-based Calibration in SBI
Local Classifier Two-Sample Tests (L-C2ST)
Model Misspecification in SBI
Visualization
Analysing variability and compensation mechanisms with conditional distributions
Plotting functionality
Active subspaces for sensitivity analysis
SBI application explorer
More guides/resources
API reference
Prior and simulator
BoxUniform
MultipleIndependent
sbi.utils.process_prior
sbi.utils.process_simulator
RestrictedPrior
RestrictionEstimator
sbi.inference.simulate_for_sbi
Neural nets
sbi.neural_nets.classifier_nn
sbi.neural_nets.likelihood_nn
sbi.neural_nets.marginal_nn
sbi.neural_nets.posterior_flow_nn
sbi.neural_nets.posterior_nn
sbi.neural_nets.posterior_score_nn
Embedding nets
CausalCNNEmbedding
CNNEmbedding
FCEmbedding
LRUEmbedding
PermutationInvariantEmbedding
ResNetEmbedding1D
ResNetEmbedding2D
SpectralConvEmbedding
TransformerEmbedding
Training
BNRE
FMPE
MarginalTrainer
MCABC
MNLE
MNPE
NLE_A
NPE_A
NPE_B
NPE_C
NPSE
NRE_A
NRE_B
NRE_C
SMCABC
Potentials
sbi.inference.likelihood_estimator_based_potential
sbi.inference.mixed_likelihood_estimator_based_potential
sbi.inference.posterior_estimator_based_potential
sbi.inference.ratio_estimator_based_potential
sbi.inference.vector_field_estimator_based_potential
Posteriors
DirectPosterior
EnsemblePosterior
ImportanceSamplingPosterior
MCMCPosterior
RejectionPosterior
VectorFieldPosterior
VIPosterior
Posterior Parameters
DirectPosteriorParameters
ImportanceSamplingPosteriorParameters
MCMCPosteriorParameters
RejectionPosteriorParameters
VectorFieldPosteriorParameters
VIPosteriorParameters
Diagnostics
sbi.diagnostics.calc_misspecification_logprob
sbi.diagnostics.calc_misspecification_mmd
sbi.diagnostics.check_sbc
sbi.diagnostics.check_tarp
sbi.diagnostics.get_nltp
LC2ST
sbi.diagnostics.run_sbc
sbi.diagnostics.run_tarp
Analysis
ActiveSubspace
sbi.analysis.conditional_corrcoeff
sbi.analysis.conditional_pairplot
sbi.analysis.conditional_potential
sbi.analysis.marginal_plot
sbi.analysis.pairplot
sbi.analysis.plot_tarp
sbi.analysis.pp_plot
sbi.analysis.pp_plot_lc2st
sbi.analysis.sbc_rank_plot
Utilities
sbi.utils.get_density_thresholder
sbi.utils.mcmc_transform
sbi.utils.transformed_potential
FAQ
What should I do when my ‘posterior samples are outside the prior support’ in SNPE?
Can the algorithms deal with invalid data, e.g., NaN or inf?
When using multiple workers, I get a pickling error. Can I still use multiprocessing?
Developer notes
How to contribute
Enhancement Proposals
Code of Conduct
Changelog
Credits
Examples
Examples
#
Inference on Hodgkin-Huxley model: tutorial
SBI for decision-making models
