botorch.fit

Model fitting routines.

botorch.fit.fit_gpytorch_mll(mll, closure=None, optimizer=None, closure_kwargs=None, optimizer_kwargs=None, **kwargs)[source]

Clearing house for fitting models passed as GPyTorch MarginalLogLikelihoods.

If a model defines a custom_fit method, it will be called directly. Otherwise, a fit method is determined based on the types of the model and MLL.

Parameters:

  • mll (MarginalLogLikelihood) – A GPyTorch MarginalLogLikelihood instance.

  • closure (Callable[[], tuple[Tensor, Sequence[Tensor | None]]] | None) – Forward-backward closure for obtaining objective values and gradients. Responsible for setting parameters’ grad attributes. If no closure is provided, one will be obtained by calling get_loss_closure_with_grads.

  • optimizer (Callable | None) – User specified optimization algorithm. When optimizer is None, this keyword argument is omitted when calling the underlying fit routine.

  • closure_kwargs (dict[str, Any] | None) – Keyword arguments passed when calling closure.

  • optimizer_kwargs (dict[str, Any] | None) – A dictionary of keyword arguments passed when calling optimizer.

  • **kwargs (Any) – Keyword arguments passed to the underlying fit routine. Unexpected keywords are ignored.

Returns:

The mll instance. If fitting succeeded, then mll will be in evaluation mode, i.e. mll.training == False. Otherwise, mll will be in training mode.

Return type:

MarginalLogLikelihood

botorch.fit.fit_fully_bayesian_model_nuts(model, max_tree_depth=6, warmup_steps=512, num_samples=256, thinning=16, disable_progbar=False, jit_compile=False, seed=0)[source]

Fit a fully Bayesian model using the No-U-Turn-Sampler (NUTS)

Uses NumPyro’s NUTS implementation (backed by JAX) for MCMC inference.

Parameters:

  • model (AbstractFullyBayesianSingleTaskGP | SaasFullyBayesianMultiTaskGP) – Fully Bayesian GP to be fitted.

  • max_tree_depth (int) – Maximum tree depth for NUTS

  • warmup_steps (int) – The number of burn-in steps for NUTS.

  • num_samples (int) – The number of MCMC samples. Note that with thinning, num_samples / thinning samples are retained.

  • thinning (int) – The amount of thinning. Every nth sample is retained.

  • disable_progbar (bool) – A boolean indicating whether to print the progress bar and diagnostics during MCMC.

  • jit_compile (bool) – Whether to use jit. Using jit may be ~2X faster (rough estimate), but it will also increase the memory usage and sometimes result in runtime errors.

  • seed (int) – Random seed for JAX PRNG.

Return type:

None

Example

>>> gp = SaasFullyBayesianSingleTaskGP(train_X, train_Y)
>>> fit_fully_bayesian_model_nuts(gp)
botorch.fit.get_fitted_map_saas_model(train_X, train_Y, train_Yvar=None, input_transform=None, outcome_transform=None, tau=None, optimizer_kwargs=None)[source]

Get a fitted MAP SAAS model with a Matern kernel.

Parameters:

  • train_X (Tensor) – Tensor of shape n x d with training inputs.

  • train_Y (Tensor) – Tensor of shape n x 1 with training targets.

  • train_Yvar (Tensor | None) – Optional tensor of shape n x 1 with observed noise, inferred if None.

  • input_transform (InputTransform | None) – An optional input transform.

  • outcome_transform (OutcomeTransform | None) – An optional outcome transform.

  • tau (Tensor | float | None) – Fixed value of the global shrinkage tau. If None, the model places a HC(0.1) prior on tau. Can be a tensor for batched models where each batch has a different sparsity prior.

  • optimizer_kwargs (dict[str, Any] | None) – A dict of options for the optimizer passed to fit_gpytorch_mll.

Returns:

A fitted SingleTaskGP with a Matern kernel.

Return type:

SingleTaskGP