botorch

botorch copied to clipboard

[Feature Request] Add an option to specify gen_candidates method in joint_optimize

8

# 🚀 Feature Request Add an argument in [joint_optimize](https://github.com/pytorch/botorch/blob/master/botorch/optim/optimize.py#L161) to specify gen_candidates method - so you can easily use e.g. `fit_gpytorch_torch` instead of `fit_gpytorch_scipy`. ## Motivation It often makes sense...

Tenoke

[Feature Request] Simplex constraint on parameters

2

# 🚀 Feature Request We'd like to support a simplex constraint on a subset of parameters (all parameters positive, sum to 1). ## Motivation See e.g. discussion in #480

Balandat

[Bug] MultiDeviceKernel not supported in Botorch

8

# 🐛 Bug Hi, I wanted to train `SingleTaskGP` on multiple GPUs as I have got 8 cards on my node. So I searched and found out about the Gpytorch's...

Bibyutatsu

Weird fitting behavior after condition_on_observations

1

Using `fit_gpytorch_model` on a model obtained using `condition_on_observations` produces bogus fits. This was described in https://github.com/pytorch/botorch/issues/337#issuecomment-563358045. Basic model: ``` import torch from torch import Tensor from botorch.models import SingleTaskGP from...

Balandat

Buxfix for Proximal acquisition function wrapper for negative base acquisition functions

7

## Motivation This PR fixes a major issue when using the ```ProximalAcquisitionFunction``` with base acquisition functions that are not strictly positive. This PR fixes it by applying a Softplus transformation...

roussel-ryan

input_transform Normalize does not seem to work properly with condition_on_observations

15

# 🐛 Bug After running `model.condition_on_observations(new_x, new_y)`, where the original model was instantiated with `Normalize(d)`, that model fails during retraining. I believe this is a bug but I'm honestly not...

matthewcarbone

Improving test coverage of UnifiedSkewNormal code

16

Summary: This commit improves the test coverage of the code located in botorch/utils/probability. For the current coverage without this commit, [see here](https://app.codecov.io/gh/pytorch/botorch/pull/1394). Differential Revision: D39556258

SebastianAment

Loss closures, fit_gpytorch_torch, fit_gyptorch_mll dispatch for ApproximateGPs

7

Summary: **Changelog:** - Enable user-defined loss closures. - `fit_gptorch_torch` rewrite - Add `fit_gyptorch_mll` dispatch for ApproximateGPs Differential Revision: D39101211

j-wilson

[Bug]Gpmodel raises an error when fitting to my input data

6

# 🐛 Bug ## To reproduce ** Code snippet to reproduce ** ```python fit_gpytorch_model(mll_ei) ``` ** Stack trace/error message ** ``` gpytorch.utils.errors.NotPSDError: Matrix not positive definite after repeatedly adding jitter...

ovr4

[Feature Request] Minibatch training when mll is an _ApproximateMarginalLogLikelihood

11

# 🚀 Feature Request Now that `fit_gpytorch_mll` exists using multiple dispatch, it seems like it'd be fairly straightforward to support minibatch training by registering a `fit_gpytorch_torch_stochastic` or similar as the...

jacobrgardner