From 664245be50a447b5d12a4ec939a7fbe7f7041b6c Mon Sep 17 00:00:00 2001 From: Sebastian Musslick Date: Mon, 2 Dec 2024 11:36:18 +0100 Subject: [PATCH] Update workflow.md naming fix --- docs/examples/closed-loop-bandit-task/workflow.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/examples/closed-loop-bandit-task/workflow.md b/docs/examples/closed-loop-bandit-task/workflow.md index 1b43f50c1..fe41997be 100644 --- a/docs/examples/closed-loop-bandit-task/workflow.md +++ b/docs/examples/closed-loop-bandit-task/workflow.md @@ -110,7 +110,7 @@ Note that it is important to use the `@on_state()` decorator to ensure that the ### Model Disagreement Sampler -The `ModelDisagreementPooler` is used to generate new experiment conditions in each cycle. It uses the best two models discovered so far to generate new reward trajectories that best distinguish between the two models. +The `model_disagreement_sampler` is used to generate new experiment conditions in each cycle. It uses the best two models discovered so far to generate new reward trajectories that best distinguish between the two models. Since the predictions of a model has a non-standard format (it isn't a single number but a 2 x ``n_trials`` array), we need to create a custom distance function.