search

microsoft / hi-ml

HI-ML toolbox for deep learning for medical imaging and Azure integration
https://aka.ms/hi-ml
MIT License
269 stars 60 forks source link

Consider not creating any SDK v2 datasets even when using V2 SDK for submission #879

Open ant0nsc opened 1 year ago

ant0nsc commented 1 year ago

Job submission with SDK v2 creates v2 data assets by default. They can't be consumed in SDK v1 jobs, and can't be mounted for local execution.

It hence would be better to always create v1 datasets, they create less problems