I think you are correct, None values should be listed as empty values not the String None.
What's the clearml version you are using? And could you retest with the latest RC?
Hi @<1643423185791619072:profile|DashingCentipede5>
Notice that you called "start_locally", it tries to run the code locally inside your jupter notebook, it assumes everything including code already exists, is that your case ?
yes, I am running the pipeline locally. I installed clearml-server and wanted to check its functionalities and was testing out optimization component
thanks Martin, I was the one who raised this bug report. I think clearml should make sure that the parameter datatypes are not being changed when loading them up. I have also proposed a solution to the ultralytics YOLO team to take care of such issue. I hope that both the companies can work on this to resolve the issue
I tried the full path a well but its still giving some error. I realized that some of the hyper parameter has None values, and these are passed as string during the tuning process , which is causing the error. I tried replacing the none values and the error is vanished. However, there are a lot of None values in the hyper-parameters and some of them are supposed to be that way. Is there something I can do about it?
I assume issue: None
Yeah this is odd I noticed as well. Let me ask the guys to take a look
which was trained on jupyter notebook.
Hmm that might be the issue, it assumes a local script running, let me verify that
Tuesday base task was a yolo model which was trained on jupyter notebook. I have used the task id of this as base-task-id
the parameter datatypes are not being changed when loading them up.
These are the auto logged parameters , inside YOLO, correct?
Just to make sure, you can actually see the value None in the UI, is that correct? (if everything works as expected, you should see empty string there)
I just updated clearml to the latest version, I still get the same error.
yes, Im trying to run the tuning in the same jupyter notebook
btw, I looked deeper into the log:
File "/tmp/tmpfa8ifmka.py", line 80, in <module>
model.train(data='coco128.yaml',epochs=20)
I'm assuming this all starts here, I think that the pipeline is Not running the code from the same folder, and you are just missing the 'coco128.yaml' try to pass a full path, wdyt?
I think I was not able to fully express my point. Let me try again.
When you are running the pipeline Fully locally (both logic and components) the assumption is this is for debugging purposes.
This means that the code of each component is locally available, could that be a reason?
I was referring the error which I was getting and realize that some of the hyperparameters that are autologged with cleaml doesnot exist in yolov8. ITs just a hunch, I might be wrong