Closed Can-Zhao closed 1 year ago
Thanks @Can-Zhao , I will add the unit tests on these 2 bundles within this PR to prevent similar issues.
Hi @Nic-Ma , when preparing the unit test of ldm bundles, I just found a place of ldm bundles that is not matched up with: https://github.com/Project-MONAI/MONAI/blob/e6ec945e4b87b90835cdad29ea64b1f27b8accda/monai/bundle/properties.py#L61
Some datasets may not have "data", such as all datasets in: https://github.com/Project-MONAI/MONAI/blob/dev/monai/apps/datasets.py or customized datasets.
Is it possible to change it into BundleProperty.REQUIRED: False
? Or ldm bundles should extend BundleWorkflow
directly and define a customized workflow (within the unit test) ?
Hi @yiheng-wang-nv ,
That's a good point, I would suggest to change the required train_dataset_data
into BundleProperty.REQUIRED: False
in the properties.
CC @SachidanandAlle and @tangy5 .
Thanks.
Hi @yiheng-wang-nv ,
That's a good point, I would suggest to change the required
train_dataset_data
intoBundleProperty.REQUIRED: False
in the properties. CC @SachidanandAlle and @tangy5 .Thanks.
I see. Hi @SachidanandAlle @tangy5 , do you think we can change the "BundleProperty.REQUIRED" of train_dataset_data
? Are there any concerns here?
/build
/build
Fixes # .
Description
update ldm noise scheduler api for inference
Status
Ready/Work in progress/Hold
Please ensure all the checkboxes:
./runtests.sh --codeformat
.version
andchangelog
inmetadata.json
if changing an existing bundle.CONTRIBUTING.md
).monai
,pytorch
andnumpy
are correct inmetadata.json
.eval_metrics
of the provided weights and TorchScript modules.large_file.yml
./home/your_name/
for"bundle_root"
).