0.3.0

Bayesian Optimization in PyTorch

Introduction
Get Started
Tutorials

Key Features

Modular

Plug in new models, acquisition functions, and optimizers.

Built on PyTorch

Easily integrate neural network modules. Native GPU & autograd support.

Scalable

Support for scalable GPs via GPyTorch. Run code on multiple devices.

References

BoTorch: Programmable Bayesian Optimization in PyTorch
@article{balandat2019botorch,
  Author = {Maximilian Balandat and Brian Karrer and Daniel R. Jiang and Samuel Daulton and Benjamin Letham and Andrew Gordon Wilson and Eytan Bakshy},
  Journal = {arXiv e-prints},
  Month = oct,
  Pages = {arXiv:1910.06403},
  Title = {{BoTorch: Programmable Bayesian Optimization in PyTorch}},
  Year = 2019}
Differentiable Expected Hypervolume Improvement for Parallel Multi-Objective Bayesian Optimization
@article{daulton2020differentiable,
  Author = {Samuel Daulton and Maximilian Balandat and Eytan Bakshy},
  Journal = {Arxiv e-prints},
  Title = {Differentiable Expected Hypervolume Improvement for Parallel Multi-Objective Bayesian Optimization},
  Year = {2020},
  url = {https://arxiv.org/abs/2006.05078}
  }

Get Started

  1. Install BoTorch:

    via conda (recommended):
    conda install botorch -c pytorch -c gpytorch
    via pip:
    pip install botorch

  2. Fit a model:

    import torch
from botorch.models import SingleTaskGP
from botorch.fit import fit_gpytorch_model
from botorch.utils import standardize
from gpytorch.mlls import ExactMarginalLogLikelihood

train_X = torch.rand(10, 2)
Y = 1 - torch.norm(train_X - 0.5, dim=-1, keepdim=True)
Y = Y + 0.1 * torch.randn_like(Y)  # add some noise
train_Y = standardize(Y)

gp = SingleTaskGP(train_X, train_Y)
mll = ExactMarginalLogLikelihood(gp.likelihood, gp)
fit_gpytorch_model(mll)

  3. Construct an acquisition function:

    from botorch.acquisition import UpperConfidenceBound

UCB = UpperConfidenceBound(gp, beta=0.1)

  4. Optimize the acquisition function:

    from botorch.optim import optimize_acqf

bounds = torch.stack([torch.zeros(2), torch.ones(2)])
candidate, acq_value = optimize_acqf(
    UCB, bounds=bounds, q=1, num_restarts=5, raw_samples=20,
)
candidate  # tensor([0.4887, 0.5063])