Reputation
Badges 1
25 × Eureka!Try:task.flush(wait_for_uploads=True)
Should do the trick π
What's the trains-server version ?
HandsomeCrow5 Seems like the right place would be in the artifacts, as a summary of the experiment (as opposed to on going reporting), is that the case?
If it is then in the Artifacts tab clicking on the artifact should open another tab with your summary, which sounds like what you were looking for (with the exception of the preview thumbnail π
Pretty confusing that neither
services
StickyLizard47 basically this is how a services queue agent should be spinned:
https://github.com/allegroai/clearml-server/blob/9b108740da21f25407bd2c59583ca1c86f8e1faa/docker/docker-compose.yml#L123
When spinning on a k8s cluster, this is a bit more complicated, as it needs to work with the clearml-k8s-glue.
See here how to spin it on k8s
https://github.com/allegroai/clearml-agent/tree/master/docker/k8s-glue
But itβs running in docker mode and it is trying to ssh into the host machine and failing
It is Not sshing to the machine it is sshing directly Into the container.
Notice the port is is sshing to is 10022 which is mapped into the container
Oh I see, this seems like Triton configuration issue, usually dim -1 means flexible. I can also mention that serving 1.1 should be released later this week with better multiple input support for triton. Does that make sense?
FYI: These days TB became the standard even for pytorch (being a stand alone package), you can actually import it from torch.
There is an example here:
https://github.com/allegroai/trains/blob/master/examples/frameworks/pytorch/pytorch_tensorboard.py
HealthyStarfish45 did you manage to solve the report_image issue ?
BTW: you also have
https://github.com/allegroai/trains/blob/master/examples/reporting/html_reporting.py
https://github.com/allegroai/trains/blob/master/examples/reporting/...
Maybe before everything else, can you share some background on the rational if starting a new sub process?
PunyBee36 to get https add an aws elb before the server , the elb will add the https to any outside connection
It would be nice to have some documentation proclaiming how randomness behaves when running tasks (in all their variations). E.g. Should I trust seeds to be reset or should I not assume anything and do my own control over seeds.
That is a good point, I'll make sure we mention it somewhere in the docs. Any thoughts on where?
Sorry, what I meant is that it is not documented anywhere that the agent should run in docker mode, hence my confusion
This is a good point! I'll make sure we stress it (BTW: it will work with elevated credentials, but probably not recommended)
Hmm I think you are correct:param auto_create: Create new dataset if it does not exist yet
it should have created it, this seems like a bug, I'll make sure to pass along π
I would recommend reading this blog post, it should give you a glimpse of what can be built π
https://medium.com/pytorch/how-trigo-built-a-scalable-ai-development-deployment-pipeline-for-frictionless-retail-b583d25d0dd
But I think this error has only appeared since I upgraded to version 1.1.4rc0
Hmm let me check something
Hi VexedCat68
Check this example:
https://github.com/allegroai/clearml/blob/4f9aaa69ed2d5b8ea68ebee5508610d0b1935d5f/examples/scheduler/trigger_example.py#L44
Hi DilapidatedDucks58
eg, we want max validation accuracy and all other metric values for the corresponding epoch
Is this the equivalent of nested sort ?
Wouldn't you get the requested behavior if you add all metric columns but sort based on the "accuracy" column ?
JitteryCoyote63 Not sure how/why the X-Pack feature was on (it is not used by the system), but you can disable it with an environment variable in the docker-composexpack.security.enabled=false
Should solve the problem ...
Hm, one of the issues I have with this change is that now every dataset hat doesnβt have a semantic version cannot be loaded anymore
Okay we definitely need to solve that.
Any chance I can ask to open a github issue (just so we do not forget).
I will pass it quickly along so that we can maybe offer a fix in the next RC
Hi SmallGiraffe94
I think it now has to be a semantic version (like pyhton packages for example)
This is so that the auto version increment can bump to the next one automatically.
Maybe adding the date as a tag would make sense? what do you think?
Or maybe in the description field
so I guess this could be one reason to start about thinking upgrading ....
Wait you mean the clearml-server ? (there is no reason not to upgrade the python package)
TenseOstrich47 every agent instance has its own venv copy. Obviously every new experiment will remove the old venv and create a new one. Make sense?
Hmm I think this was the fix (only with TF2.4), let me check a sec
right click on the experiment, select Reset, now you can edit it.
From the docs I think what's going on is that the https://opennmt.net/OpenNMT-tf/package/opennmt.Runner.html#opennmt.Runner.train is spinning a new subprocess, and the training itself happens on the subprocess.
If this is the case this will explain the lack of automagic, as the subprocess is lacking the "Task.init" call
wdyt, could that be the case ?
AstonishingWorm64 can you share the full log (In the UI under Results/Console there is a download button)?
This is the reason you are getting an error π
Basically the session asks the agent to setup a new SSH server with credentials on the remote machine, this is not an issue inside a container, as this is an isolated environment, but when running in venv mode the User running the agent is not root, hence it cannot spin/configure an SSH server.
Make sense ?
Is it being used to ssh to the instance?
It is used for the SSH client so it "knows" the SSH server (does that make sense) ?
LovelyHamster1 Now I see... Interesting credentials ability. Specifically all the S3 access on trains is derived from the ~/clearml.conf
credentials section :
https://github.com/allegroai/clearml/blob/ebc0733357ac9ead044d0ed32d41447763f5797e/docs/clearml.conf#L73
( or the AWS S3 environment variables )
I'm not sure how this AWS feature works, I suspect it is changing the AWS_ACCESS_KEY_ID AWS_SECRET_ACCESS_KEY variables on the ec2 instance. If this is the case, it should work out of...
the Task scheduler itself is a Task. What we did is we added a new parameter section on the Task (the task.connect call), so that we can later clone and modify it and use the new value in runtime
(Task.connect will put the data from the Task/UI back into the dict when the agent is running the Scheduler)
Does that make sense?