Skip to content

[Auto3DSeg] Update ALGO_BASH for support num_worker and test cases to reflect unification of epoch-based training#5763

Merged
wyli merged 4 commits intoProject-MONAI:devfrom
dongyang0122:auto3dseg_update_test_cases_v1
Dec 19, 2022
Merged

[Auto3DSeg] Update ALGO_BASH for support num_worker and test cases to reflect unification of epoch-based training#5763
wyli merged 4 commits intoProject-MONAI:devfrom
dongyang0122:auto3dseg_update_test_cases_v1

Conversation

@dongyang0122
Copy link
Copy Markdown
Collaborator

@dongyang0122 dongyang0122 commented Dec 16, 2022

Signed-off-by: dongy [email protected]

Description

  • Update ALGO_HASH to support num_workers override
  • Update test cases.

Types of changes

  • Non-breaking change (fix or new feature that would not break existing functionality).
  • New tests added to cover the changes.
  • Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
  • Quick tests passed locally by running ./runtests.sh --quick --unittests --disttests.

Signed-off-by: dongy <[email protected]>
@mingxin-zheng
Copy link
Copy Markdown
Contributor

@dongyang0122 can you also update hpo_param in tests/test_integration_autorunner.py?

hpo_param = {
    "CUDA_VISIBLE_DEVICES": train_param["CUDA_VISIBLE_DEVICES"],
    "num_epochs_per_validation": train_param["num_epochs_per_validation"],
    "num_images_per_batch": train_param["num_images_per_batch"],
    "num_epochs": train_param["num_epochs"],
    "use_pretrain": train_param["use_pretrain"],
    "pretrained_path": train_param["pretrained_path"],
    # below are to shorten the time for dints
    "training#num_epochs_per_validation": train_param["num_epochs_per_validation"],
    "training#num_images_per_batch": train_param["num_images_per_batch"],
    "training#num_epochs": train_param["num_epochs"],
    "searching#num_epochs_per_validation": train_param["num_epochs_per_validation"],
    "searching#num_images_per_batch": train_param["num_images_per_batch"],
    "searching#num_epochs": train_param["num_epochs"],
    "searching#num_warmup_iterations": train_param["num_warmup_epochs"],
    "nni_dry_run": True,
}

Signed-off-by: dongy <[email protected]>
@dongyang0122
Copy link
Copy Markdown
Collaborator Author

@dongyang0122 can you also update hpo_param in tests/test_integration_autorunner.py?

hpo_param = {
    "CUDA_VISIBLE_DEVICES": train_param["CUDA_VISIBLE_DEVICES"],
    "num_epochs_per_validation": train_param["num_epochs_per_validation"],
    "num_images_per_batch": train_param["num_images_per_batch"],
    "num_epochs": train_param["num_epochs"],
    "use_pretrain": train_param["use_pretrain"],
    "pretrained_path": train_param["pretrained_path"],
    # below are to shorten the time for dints
    "training#num_epochs_per_validation": train_param["num_epochs_per_validation"],
    "training#num_images_per_batch": train_param["num_images_per_batch"],
    "training#num_epochs": train_param["num_epochs"],
    "searching#num_epochs_per_validation": train_param["num_epochs_per_validation"],
    "searching#num_images_per_batch": train_param["num_images_per_batch"],
    "searching#num_epochs": train_param["num_epochs"],
    "searching#num_warmup_iterations": train_param["num_warmup_epochs"],
    "nni_dry_run": True,
}

Thanks for the suggestions! I have updated this part.

@mingxin-zheng mingxin-zheng changed the title [auto3dseg] update_test_cases [Auto3DSeg] Update ALGO_BASH for support num_worker and test cases to reflect unification of epoch-based training Dec 19, 2022
@wyli
Copy link
Copy Markdown
Contributor

wyli commented Dec 19, 2022

/build

@wyli wyli enabled auto-merge (squash) December 19, 2022 09:21
@wyli wyli merged commit 8037fcd into Project-MONAI:dev Dec 19, 2022
@dongyang0122 dongyang0122 deleted the auto3dseg_update_test_cases_v1 branch February 22, 2023 15:04
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants