Updates optimas to use the latest libEnsemble release.
Highlights
Use new libEnsemble interface for specifying resources (set num_procs and num_gpus in gen_specs["out"]). This is more portable than the previous method and will work both with NVIDIA and AMD GPUs.
All the output of an optimas Exploration (including the logs) is now stored in the exploration folder. This is done using the new workflow_dir feature of libEnsemble. The simulation folders are located in exploration/evaluations.
Use new libEnsemble interface for specifying resources (set num_procs and num_gpus in gen_specs["out"]). This is more portable than the previous method in which we were setting CUDA_VISIBLE_DEVICES.
Replace set_env_to_slots with set_env_to_gpus in the generator function.
The generator uses resources even when gen_resources is set to 0. To make sure no resources are allocated for the generator, we now set libE_specs['zero_resource_workers'] = [1]
All the output of an optimas Exploration (including the logs) is now stored in the exploration folder. This is done using the new workflow_dir feature of libEnsemble. The simulation folders are located in exploration/evaluations.
Reduce number of simulation workers to 2 in tests so that they can run properly on GitHub actions (the new libEnsemble will raise an error if we have more workers than CPUs).
Adds a new _reset_libensemble method to the Exploration that makes sure that some module variables in libEnsemble are reset after finishing the run. This is needed in order to launch several runs from the same script.
The paths to the sim_dir_copy_files are now made absolute before passing them to libEnsemble.
n_proc renamed to n_procs in the TemplateEvaluator.
By default n_gpus=0 (it used to be 1).
Increase version to 0.2.0.
From the changes above, these will affect the user:
All the output is now stored in the exploration directory, with the simulation folders being in exploration/evaluations.
n_proc renamed to n_procs in the TemplateEvaluator. (this does not really have an impact, as n_proc was just a placehold with no effect until now)
By default n_gpus=0 (it used to be 1).
An exploration run will now fail to start if more resources than available are required by the simulations (e.g., more workers than CPUs). This comes from the new libEnsemble resource asignement.
Updates
optimasto use the latestlibEnsemblerelease.Highlights
num_procsandnum_gpusingen_specs["out"]). This is more portable than the previous method and will work both with NVIDIA and AMD GPUs.Exploration(including the logs) is now stored in theexplorationfolder. This is done using the newworkflow_dirfeature of libEnsemble. The simulation folders are located inexploration/evaluations.Note
It currently needs the develop libE branch.
Do not merge until these fixes have been included in a libEnsemble release:
Before merging:
Changes
num_procsandnum_gpusingen_specs["out"]). This is more portable than the previous method in which we were settingCUDA_VISIBLE_DEVICES.set_env_to_slotswithset_env_to_gpusin the generator function.gen_resourcesis set to0. To make sure no resources are allocated for the generator, we now setlibE_specs['zero_resource_workers'] = [1]Exploration(including the logs) is now stored in theexplorationfolder. This is done using the newworkflow_dirfeature of libEnsemble. The simulation folders are located inexploration/evaluations._reset_libensemblemethod to theExplorationthat makes sure that some module variables in libEnsemble are reset after finishing the run. This is needed in order to launch several runs from the same script.sim_dir_copy_filesare now made absolute before passing them to libEnsemble.n_procrenamed ton_procsin theTemplateEvaluator.n_gpus=0(it used to be1).0.2.0.From the changes above, these will affect the user:
explorationdirectory, with the simulation folders being inexploration/evaluations.n_procrenamed ton_procsin theTemplateEvaluator. (this does not really have an impact, asn_procwas just a placehold with no effect until now)n_gpus=0(it used to be1).