Sait Cakmak

Fix torchx import error

6

Summary: `from ax.runners.torchx import TORCHX_TRACKER_BASE` leads to an import error in Python 3.7 unit tests. Moving this into try/except block to get around the issue. Example failing run https://github.com/facebook/Ax/runs/6891472698?check_suite_focus=true Reviewed...

Additional acquisition functions for nightly benchmark

1

Summary: Adds a few single-objective acquisition functions (that work out-of-box) to the nightly benchmark run. Reviewed By: Balandat Differential Revision: D37087388

Update the `publish_site` docs to unblock website deployment

4

The existing `publish_site` script attempts to force-push to the gh-pages branch, which gets rejected due to gh-pages being a protected branch. This updates the script to modify the git repo...

[Bug] model.cpu() does not move the caches, leading to device error

4

# 🐛 Bug Calling `model.cpu()` on a `model` that is originally created on the GPU does not move the caches (e.g., `mean_cache`) to the CPU leading to a `RuntimeError` due...

[Feature Request] Docs listing available models and their properties

2

# 🚀 Feature Request I find myself searching through the docstrings or the API reference whenever I am not sure what model to use. These typically have enough information to...

SingleTaskGP with empty inputs [Do not land]

3

Summary: See https://github.com/pytorch/botorch/discussions/1217 Differential Revision: D36490304

[RFC] Refactor Input Transforms

3

Summary: Currently, we apply the input transforms in `train` mode at the `forward` call, and in `eval` model at the `posterior` call. We also use a `transform_train_inputs` call at the...

Update tutorials to use "sample_around_best": True

8

## Motivation Using `"sample_around_best": True` should lead to improved optimization performance. ### Have you read the [Contributing Guidelines on pull requests](https://github.com/pytorch/botorch/blob/main/CONTRIBUTING.md#pull-requests)? Yest ## Test Plan TODO: Re-run the notebooks, so...

Questions regarding contextual GPs / LCEMGP

17

I want to use the contextual GPs, in particular the `LCEMGP`, for a project. The existing documentation, i.e., the docstrings, do not present any examples, which makes it challenging to...

[RFC] Move input transforms to GPyTorch

1

This diff presents a minimal implementation of input transforms in GPyTorch, as requested in #1652. This should be viewed together with pytorch/botorch#1372. The input transforms themselves are currently implemented in...