The same training works sometimes. But I'm not sure how to troubleshoot when it stops logging the metrics
I'm not sure how to even troubleshoot this.
Is there someway to kill all connections of a machine to the ClearML server this does seem to be related to restarting a task / running a new task quickly after a task fails or is aborted
That makes sense... If you turn auto_connect_streams to false this mean that auto reporting will be disabled as per the documentation.. If you turn it to True then logging should resume.
It is still getting stuck. I think the issue might have something to do with the iterations versus epochs. I notice that one of the scalars that gets logged early is logging the epoch while the remaining scalars seem to be iterations because the iteration value is 1355 instead of 26
Not sure why that is related to saving images
Hi @<1719524641879363584:profile|ThankfulClams64>
I am using ClearML Pro and pretty regularly I will restart an experiment and nothing will get logged to ClearML.
I use ClearML with pytorch 1.7.1, pytorch-lightning 1.2.2 and Tensorboard auto
All ClearML has the latest stable updates. (clearml 1.7.4, clearml-agent 1.7.2)
Is this still happening with the latest clearml ( clearml==1.16.3rc2
) ?
What is the TB version?
I remember a fix regrading lightining support
Also just making sure, are you using the default lightning TB logger ?
How are you initializing the Task.init
(i.e. could you copy here the code?)
Okay I will do another run to capture the console output. We currently set auto_connect_streams to False to reduce the number of API calls. So there isn't really anything in the ClearML task page console section
This was on the same machine I am having issues with it logs scalars correctly using the example code, but when I add in that callback which just logs a random image to tensorboard I don't get any scalars logged
It was working for me. Anyway I modified the callback. Attached is the script that has the issue for me whenever I add random_image_logger
to the callbacks It only logs some of the scalars for 1 epoch. It then is stuck and never recovers. When I remove random_image_logger
the scalars are correctly logged. Again this only on 1 computer, other computers we have logging work perfectly fine
I'll update my clearml version. Unfortunately I do not have a small code snippet and it is not always repeatable. Is there some additional logging that can be turned on?
If you remove any reference of ClearML from the code on that machine, does it still hang?
I am on 1.16.2
task = Task.init(project_name=model_config['ClearML']['project_name'],
task_name=model_config['ClearML']['task_name'],
continue_last_task=False,
auto_connect_streams=True)
My bad, if you set auto_connect_streams to false, you basically disable the console logging... Please see the documentation:
auto_connect_streams (Union[bool, Mapping[str, bool]]) – Control the automatic logging of stdout and stderr.
Yes it shows on the UI and has the first epoch for some of the metrics but that's it. It has run like 50 epochs, it says it is still running but there are no updates to the scalars or debug samples
So even if you abort it on the start of the experiment it will keep running and reporting logs?
It seems similar to this None is it possible saving too many model weights causes metric logging thread to die?
Then we also connect two dictionaries for configs
task.connect(model_config)
task.connect(DataAugConfig)
It is not always reproducible it seems like something that we do not understand happens then the machine consistently has this issue. We believe it has something to do with stopping and starting experiments
Yea I am fine not having the console logging. My issues is the scalers and debug images occasionally don't record to ClearML
@<1719524641879363584:profile|ThankfulClams64> , can you provide a small code snippet that reproduces this behaviour? Can you also test with the latest version of clearml
?
I am using 1.15.0. Yes I can try with auto_connect_streams set to True I believe I will still have the issue
@<1719524641879363584:profile|ThankfulClams64> , if you set auto_connect_streams to false nothing will be reported from your frameworks. With what frameworks are you working, tensorboard?
Hi we are currently having the issue. There is nothing in the console regarding ClearML besides
ClearML Task: created new task id=0174d5b9d7164f47bd10484fd268e3ff
======> WARNING! Git diff too large to store (3611kb), skipping uncommitted changes <======
ClearML results page:
The console logs continue to come in put no scalers or debug images show up.
I am still having this issue. An update is that the "abort" does not work. Even though the state is correctly tracked in ClearML when I try to abort the experiment through the UI it says it does it but the experiment remains running on the computer.
Thanks @<1719524641879363584:profile|ThankfulClams64> having a code that can reproduce it is exactly what we need.
One thing I might have missed and is very important , what is your tensorboard package version?
There is clearly some connection to the ClearML server as it remains "running" the entire training session but there are no metrics or debug samples. And I see nothing in the logs to indicate there is an issue