botorch
botorch copied to clipboard
pytorch
Bayesian optimization in PyTorch
Hi everyone, I have developed a custom Multi-Layer Perceptron (MLP) model using PyTorch, which has been pre-trained for my specific application. The model architecture has an input size of 579...
Hi, I have been following the tutorial [VAE for MNIST in BoTorch](https://botorch.org/tutorials/vae_mnist) and was wondering if it's possible to replace the combination of the VAE and single-task GP model with...
Summary: Adds a helper for evaluating the feasibility of intra-point parameter constraints on a given tensor. Differential Revision: D63909338
# 🐛 Bug Unable to fit a Heteroskedastic GP with Warp input transform. ## To reproduce ** Code snippet to reproduce ** ```python # Define some training data as a...
# 🐛 Bug When trying to use `batch_cross_validation` to fit a `KroneckerMultiTaskGP`, a batch shape error occurs. The code snippet below uses a similar example to https://botorch.org/tutorials/batch_mode_cross_validation. The following snippet...
# 🐛 Bug `botorch.models.utils.gpytorch_modules` implements a few utility functions that return kernels and likelihoods. Those functions should enforce constraints on `kernel.lengthscale` and `likelihood.noise` to make sure they are always positive....
Differential Revision: D62993873
# 🐛 Bug The inputs `X` are not transformed by the model's input transform when running `GPyTorchModel.condition_on_observations(X, y)`. The outcomes `Y`, however, are transformed. This means that the new observations...
Previously, this tutorial used `normalize` & `unnormalize` helpers to manually transform the inputs before feeding them to the model & acquisition functions. This requires transforming the data in many places...
# 🚀 Feature Request I would like to define a custom q sample shape when instantiating Q acquisition function without knowing the posterior. ## Motivation The MCSamplerMixin allows the use...
