How to use ChainedOutcomeTransform with Standardize?

[samuelkim16]

I'd like to use multiple outcome transforms. A snippet of my code looks like this:

import botorch.models.transforms as transforms

outcome_transform = transforms.outcome.ChainedOutcomeTransform(
    tf1=transforms.outcome.Log(),
    tf2=transforms.outcome.Standardize(m=1),
)

model = SingleTaskGP(train_X, train_Y,
    input_transform=input_transform,
    outcome_transform=outcome_transform, 
)

However, during my optimization run, I will regularly get warnings in each iteration like the following:

WARNING:py.warnings:InputDataWarning: Data (outcome observations) is not standardized (std = tensor([31.5549], dtype=torch.float64), mean = tensor([-1.2118], dtype=torch.float64)).Please consider scaling the input to zero mean and unit variance.
 C:\Users\username\AppData\Roaming\Python\Python310\site-packages\botorch\models\utils\assorted.py: 267

Am I chaining the outcome transforms incorrectly? The warning does not give a full stack trace so it is difficult for me to troubleshoot. I do not get any errors when using Standardize alone (without ChainedOutcomeTransform) or when just initializing the model.

[saitcakmak]

Hi @samuelkim16. That's strange. I don't see anything wrong with the snippet you shared. When testing it, I see that the outcomes are correctly transformed in SingleTaskGP.__init__ before the validate_input_scaling call (where the warning would come from).

Could you share an end-to-end runnable example (with dummy data as necessary) that reproduces the issue? If you wanted to debug it a bit more closely, you could turn the warning into error using a block like this one to inspect where it comes from:

import warnings
from botorch.exceptions.warnings import InputDataWarning

with warnings.catch_warnings():
    warnings.filterwarnings("error", category=InputDataWarning)
    # execute code as usual here

[Balandat]

This sounds like a bug - made an issue out of this: #2812 - let's continue on that issue.