From 3d870c2975104e96e96b3aa5d5b18f92c2e21b01 Mon Sep 17 00:00:00 2001 From: karibbov Date: Fri, 23 Feb 2024 17:02:26 +0100 Subject: [PATCH] Deepgp, dpl, parallel runs comment --- neps/optimizers/bayesian_optimization/models/DPL.py | 4 ++++ neps/optimizers/bayesian_optimization/models/deepGP.py | 4 ++++ 2 files changed, 8 insertions(+) diff --git a/neps/optimizers/bayesian_optimization/models/DPL.py b/neps/optimizers/bayesian_optimization/models/DPL.py index 911e3486..82bd78c1 100644 --- a/neps/optimizers/bayesian_optimization/models/DPL.py +++ b/neps/optimizers/bayesian_optimization/models/DPL.py @@ -224,6 +224,10 @@ def __init__( # IMPORTANT: For parallel runs lock the checkpoint file during the whole training checkpointing: bool = False, root_directory: Path | str | None = None, + # IMPORTANT: For parallel runs use a different checkpoint_file name for each + # IMPORTANT: surrogate. This makes sure that parallel runs don't override each + # IMPORTANT: others saved checkpoint. Although they will still have some conflicts due to + # IMPORTANT: global optimizer step tracking checkpoint_file: Path | str = "surrogate_checkpoint.pth", refine_epochs: int = default_refine_epochs, n_initial_full_trainings: int = default_n_initial_full_trainings, diff --git a/neps/optimizers/bayesian_optimization/models/deepGP.py b/neps/optimizers/bayesian_optimization/models/deepGP.py index cbc6d5a9..862f8a6e 100644 --- a/neps/optimizers/bayesian_optimization/models/deepGP.py +++ b/neps/optimizers/bayesian_optimization/models/deepGP.py @@ -164,6 +164,10 @@ def __init__( # IMPORTANT: hence, it is not suitable for multiprocessing settings checkpointing: bool = False, root_directory: Path | str | None = None, + # IMPORTANT: For parallel runs use a different checkpoint_file name for each + # IMPORTANT: surrogate. This makes sure that parallel runs don't override each + # IMPORTANT: others saved checkpoint. Although they will still have some conflicts due to + # IMPORTANT: global optimizer step tracking checkpoint_file: Path | str = "surrogate_checkpoint.pth", refine_epochs: int = 50, n_initial_full_trainings: int = 10,