UnevenDolphin73 : From which URL is your most recent screenshot?
But, I guess CostlyOstrich36 wrote that in a different chat, right?
It should store it on the fileserver, perhaps you're missing a configuration option somewhere?
It is documented at None ... super deep in the code. If you don't know that output_uri in TASK's (!) init is relevant, you would never know...
Heh, well, John wrote that in the first reply in this thread 🙂
And in Task.init main documentation page (nowhere near the code), it says the following -
FWIW It’s also listed in other places VivaciousBadger56 , e.g. None says:
In order to make sure we also automatically upload the model snapshot (instead of saving its local path), we need to pass a storage location for the model files to be uploaded to.
For example, upload all snapshots to an S3 bucket…
Do you mean "exactly" as in "you finally got it" or in the sense of "yes, that was easy to miss"?
VivaciousBadger56 I'm not sure I'm following you - is the issue not being able to upload to the ClearML server or to load the downloaded file?
We'll try to add referenced to that in other places as well 👍
FWIW, we prefer to set it in the agent’s configuration file, then it’s all automatic
Yes, you're correct, I misread the exception.
Maybe it hasn't completed uploading? At least for Datasets one needs to explicitly wait IIRC
VivaciousBadger56 regrading: None
Is this a discussion or PR ?
(general ranting is saved for our slack channel 🙂 )
By the way, output_uri is also documented as part of the Task.init() docstring ( None )
UnevenDolphin73 : Thanks, but it does not mention the File Storage of "ClearML Hosted Server".
The documentation is messy, I’ve complained about it the in the past too 🙈
Hi all, sorry for not being so responsive today 🙏
I can only say I’ve found ClearML to be very helpful, even given the documentation issue.
I think they’ve been working on upgrading it for a while, hopefully something new comes out soon.
Maybe AgitatedDove14 has further info 🙂
I have already been trying to contribute (have three pull requests), but honestly I feel it is a bit weird, that I need to update a documentation about something I do not understand, while I actually try to evaluate if ClearML is the right tool for our company...
We're certainly working hard on improving the documentation (and I do apologize for the frustrating experience)
UnevenDolphin73 : I do not get this impression, because during update_weights I get the message
2023-02-21 13:54:49,185 - clearml.model - INFO - No output storage destination defined, registering local model C:\Users..._Demodaten_FF_2023-02-21_13-53-51.624362.model
Well you could start by setting the output_uri to True in Task.init .
SuccessfulKoala55 : I referenced this conversation in the issue None
Hi VivaciousBadger56 , you can configure Task.init(..., output_uri=True) and this will save the models to the clearml file server
Heh, good VivaciousBadger56 😁
I was just repeating what CostlyOstrich36 suggested, credits to him