Summary: 72 instances, 66 unique

Text	Count
# TODO (jej): Replace with inferred noise before making perf judgements.	1
It should not be used for TODO (another common use case of NIE).	1
try:  # TODO T61331534: revert this; just a hotfix for AutoML	1
acqf = Acquisition(  # TODO: For multi-fidelity, might need diff. class.	1
# TODO: Add bounds for optimization stability - requires revamp upstream	1
# TODO: Add this if all acq. functions accept the `subset_model`	1
# TODO: Allow failure behavior based on a survival rate	1
# TODO: Use metrics-like Data type field in Data instead.	1
# TODO: maybe return a copy here to guard against implicit changes	1
# Extract acquisition value (TODO: Make this less painful and repetitive)	1
# TODO: Allow passing down `covar_module_class`, `covar_module_options`,	1
# TODO: pass array_model_gen_args to _model_gen	1
# TODO: In the future, do not need to fail here; can add a "unifying" root	1
# TODO: remove rows from <values>[method] of length different	1
# TODO: Better logic for deciding when to use a ModelListGP. Currently the	1
# TODO: Create a `fit_botorch_model` method that handles the fitting.	1
# TODO: check size and add warning for large tensors: T69137799	1
# TODO: Currently, the timestamps generated below must exactly match the	1
# TODO (jej): replace dropdown with two dropdowns, one for x one for y.	1
# TODO: Check if model can predict in favor of try/catch.	1
# TODO: test alternative outputscale priors	1
# TODO: (jej) T64002590 Let this serve as a starting point for optimization.	1
obsf.parameters[p_name] = int(round(param))  # TODO: T41938776	1
raise NotImplementedError  # TODO (done in D18009570)	1
# TODO (ref: https://fburl.com/diff/uneqb3n9)	1
# TODO: Add optional timeout between retries of `run_trial(s)`.	1
# TODO: support batch evaluation	1
# @TODO T40555279: metric --> metric_name everywhere in plotting	1
optimization problems. % TODO: refer to an example.	1
# TODO: extract data type from SQAData after DataRegistry added.	1
# TODO: Allow normalizing progr_key (e.g. subtract min time stamp)	1
# TODO: Allow passing of additional kwargs to `interpolate`	1
# use_saas is deprecated. TODO: remove	4
# TODO T38563759: Sort parameters by feature importances	1
# TODO: Add commented out methods when they are brought back to modular BotAx	1
# TODO: use some container down the road (similar to	1
# TODO: Is this equivalent an "else:" here?	1
# TODO: update optimizers to handle inequality_constraints	1
# TODO: We currently only pass in `covar_module` and `likelihood` if they are	1
# TODO: implement `_get_hypervolume_trace()`	1
- TODO: multi-objective optimization	1
# TODO: move this reuseable function and its  equivalent reverse functions	1
# TODO: When we move `botorch_modular` directory to OSS, we will extend	1
# TODO replace with sobol	1
TODO: use gpytorch `Distance` module. This will require some care to make sure	1
# TODO: Remove best_point_recommender for botorch_moo. Used in modelbridge._gen.	2
# TODO: If `evaluate_suggested` is True on the problem	1
# TODO: make optional when BoTorch model factory is checked in.	1
# TODO: Figure out if there's a better way to color code out-of-sample points	1
# TODO (jej): Transform covariances.	2
# TODO: Verify whether 0, 1 weights cause problems because of subset_model.	1
# TODO: Consider parallelizing evaluation of large batches	1
# TODO (jej): rewrite optimize_acqf wrappers to avoid duplicate code.	1
- TODO: ChoiceParameter plots	1
# TODO: Allow Posterior to (optionally) return the full covariance matrix	2
# TODO: Support covariances for a subset of observations	1
# TODO: Improve performance / make less ad hoc by using a priority queue	1
return visited  # TODO: Should there be other validation?	1
# TODO: Allow using an arbitrary prediction model for this instead	1
# TODO: read pinned Botorch version from a shared source	1
# TODO: Remove once https://github.com/pytorch/pytorch/issues/41489 is resolved.	1
# TODO: update optimizers to handle inequality_constraints	1
# TODO: It will be more memory-efficient to do this filtering before	1
# TODO: Annotate and add `register_transform`	1
TODO: move some of this into botorch.	1
# TODO: The status on the experiment does not distinguish between	1