Nicholas Geneva
Nicholas Geneva
/blossom-ci
Hi all, I encountered this issue as well. Seems this occurs when parsing for workspace packages. Would there ever be an instance where a list would be present for a...
xref: https://github.com/h5py/h5py/issues/2726
Appears to be some miss match between the recent releases of netcdf4 and h5py https://github.com/Unidata/netcdf4-python/issues/1438
We will be planning to add an example that shows how one could build their own workflow that checkpoints the models state that will allow one to restart inference from...
Reposting from the h5py repo here vis, incase there's any useful information. I am encountering the similar issue. I'm using a uv environment on linux, the following produces an error...
@Steven04453X Are you just trying to run inference? Consider looking at Earth2Studio for some automated methods that can make running the model easier: - https://nvidia.github.io/earth2studio/examples/09_stormcast_example.html#sphx-glr-examples-09-stormcast-example-py - https://nvidia.github.io/earth2studio/examples/10_stormcast_ensemble_example.html#sphx-glr-examples-10-stormcast-ensemble-example-py
Regardless: I think you may be trying to load a training checkpoint in the provided script, not a model checkpoint. Try using the module load. E.g. https://github.com/NVIDIA/earth2studio/blob/main/earth2studio/models/px/stormcast.py#L261
@CharlelieLrt you're correct, please do not add to the run.py. depending on what the workflow is, maybe this should be added as an example. Regardless, I think it would be...
Thanks for opening this @sunjingan !