
Reputation
Badges 1
26 × Eureka!When creating new datasets by combinging and modifying existing datasets programmatically in python, if I add the previous datasets as "parents" I can see a nice flow graph in ClearML of how each child dataset was generated. However this means my datasets contain the content of their parents which is quite a lot, is there a way for me to have that flow graph without using the parent datasets option?
 so I can create an image with the model. Whats the recommended way of doing that?
ahh just use one of the community ones?
Ok it works, but I don't see the labels in the model output, is there a way for me to use OutputModel to update those labels?
clearml-data create --project "PROJECT" --name "NAME"
ok so:
you recommend just saving the dataset id as part of the task configuration? I think I was a bit unclear, my question is how should I report them from the code, they are not caught automatically because they are custom parameters I calculate not as part of any framework, so I wonder if I should report them as artifacts, or maybe scalars? my issue with scalars is that I only have 1 of each type, and the API seems to be oriented toward a series of results of the same type
anyway I can catch that token so I can use it directly? I looked throught he documentation and found no way to access it
(I tested this, switched it off and the datasets disappear, switch it on and they appear)
Thanks, found it, but yeah, it would be a lot more convenient if you can add it to regular right click menu the same way it is with models and tasks
yes we saved those in the hyper parameters
and yes I meant the AMI, in your docs you recommend to use the old one until you post a new one, but the old one is no longer available.
ok, a great change, but why are they empty for me? 🙂
I would like to recreate an experiment after saving its configurations, to do that that I need to load those configurations in another notebook, right now the only way I managed to do that is by saving those configurations as an artifact and load that artifact, but it is less convenient than loading a configuration.