Why does ClearML hide the dataset task from the main WebUI?
Basically you have the details from the Dataset page, why should it be mixed with the others ?
If I specified a project for the dataset, I specifically want it there, in that project, not hidden away in some
.datasets
hidden sub-project.
This maybe a request for "Dataset" tab under project, why would you need the Dataset Task itself is the main question?
Not all dataset objects are equal, and perhap...
If there is new issue will let you know in the new thread
Thanks! I would really like to understand what is the correct configuration
This one should work:
` path = task.connect_configuration(path, name=name)
if task.running_locally():
my_params = read_from_path(path)
my_params = change_parmas(my_params) # change some staff
store back the change, my_params assumed to be the content of the param file (text)
task.set_configuration_object(name=name, config_taxt=my_params) `
Well it is there, do you have it in your docker-compose as well?
https://github.com/allegroai/trains-server/blob/master/docker-compose.yml#L55
Nope - confirmed to be running on the OS's Python environment,
okay so bare metal root is definitely not recommended.
I'm not sure how/why it get's stuck though 😞
Any chance you can run the agent as non-root?
Also maybe preferred in docker mode, so it is easier for you to control the environment of the Task
Our server is deployed on a kube cluster. I'm not too clear on how Helm charts etc.
The only thing that I can think of is that something is not right the the load balancer on the server so maybe some requests coming from an instance on the cluster are blocked ...
Hmm, saying that aloud that actually could be?! Try to add the following line to the end of the clearml.conf on the machine running the agent:
api.http.default_method: "put"
Yes. Because my old
has never been resolved (though closed), we use the dataset object to upload e.g. local files needed for remote execution.
Ohh No I remember... following this line, can I assume these files are reused, i.e. this is not a "per instance" . I have to admit that I have a feeling this is a very unique usecase. and Maybe the "old" way Dataset were shown is better suited ?
No, I mean why does it show up in the task view (see attached image), forcing me to clic...
we can add non-clearml code as a step in the pipeline controller.
Yes 🙂 , btw you can kind of already do that, with pre/post function callbacks (notice they are running from the same scope as the actual pipeline controller).
What exactly did you have in mind to put there ?
Hi SteadySeagull18
However, it seems to be entirely hanging here in the "Running" state.
Did you set a an agent to listen to the "services" queue ?
Someone needs to run the pipeline logic itself, it is sometimes part of the clearml-server deployment but not a mist
FlatOctopus65
In my local environment
pipeline_package
is installed in development mode
In order to install the package you need to specify the git repo of the package, this is how the pipeline would know where to bring it from.
Either install it locally with "pip install git+ https://github.com/ ...." or add tp the packages
argument of the Pipeline wrapper packages = ["git+
https://github.com/
"] `
wdyt?
Hi @<1566596960691949568:profile|UpsetWalrus59>
just wondering - shouldn't the job still work if I didn't push the commit yet
How would that work? it does not know which commit to take? it would also fail on git diff apply, no?
Hi @<1523701066867150848:profile|JitteryCoyote63>
I found a memory leak
in
Logger.report_matplotlib_figure
Are you sure this is not Matplotlib leak but the Logger's fault ? I'm trying to think how we could create such a mem leak
wdyt?
Hi SubstantialElk6
We can't seem to find a way for the end user to pass in their git credentials when they run their codes in both agent and non-agent scenarios. Any advice here?
The bottom line is the agent needs to have read-only access to all the repositories so it can launch any Task. I would recommend to create an agent git user with read-only credentials and configure the agent to use it. wdyt?
How are you starting the agent?
Please let me know what you find 🤞
Can you try to set this in your clearml.conf:
agent.pip_download_cache.enabled: false
this should disable the local caching, of your wheel, I suspect there is some issue with the local cache file in windows...
So you mean 1.3.1 should fix this bug?
Yes it should see the release notes, there are a few "disappearing" UI fixes:
https://github.com/allegroai/clearml-server/releases/tag/v1.3.0
We could use our 8xA100 as 8 workers, for 8 single-gpu jobs running faster than on a single 1xV100 each.
@<1546665634195050496:profile|SolidGoose91> I think that in order to have the flexibility there you need the "dynamic" GPU allocation that is only part of the "enterprise" offering 😞
That said, why not allocate a single a100 machine? no?
We have tried to manually restart tasks reloading all the scalars from a dead task and loading latest saved torch model.
Hi ThickKitten19
how did you try to restart them ? how are you monitoring dying instances ? where . how they are running?
"what's the trains/trains-agent/trains-server versions ?" how can I check it?
trains/trains-agent are pip packages os,pip freeze | grep trains
trains-server you can check in the /profile page top left corner
I had again the same problem but within a remote pipeline setup.
Are you saying the ussue is not fixed? can you verify the pipeline & pipeline components are using the at least 1.104rc0 version?
Notice there is no need to upgrade the server, only the ClearML python package
I see now.
Let's assume you know which snapshot that was:
` prev_task = Task.get_task(task_id='the_first_training_task_id')
get the second from last checkpoint
task.models['output'][-2].url
prev_scalars = prev_task.get_reported_scalars()
new_task = Task.init('example', 'new task')
logger = new_task.get_logger()
do some fpr loop and report the prev_scalars with logger.report_scalars
new_task.flush(wait_for_uploads=True)
new_task.set_initial_iteration(22000)
start the train `
The current implementation (since 1.6.3 I think) creates the issues in the linked comment (with images to visualize).
Understood, basically the moment we add nested project view to the dataset (and pipelines for that matter, and both are already being worked on), it should solve everything. Is that correct?
okay that makes sense, if this is the case I would just use clearml-agent execute --id <task_id here>
to continue the training Task.
Do notice you have to reload your last chekcpoint from the Task's models/artifacts to continue 🙂
Last question, what is the HPO optimization algorithm, is it just grid/random search or optuna hbop/optuna, if this is the later, how do make it "continue" ?
Hi FreshBat85clearml_agent: ERROR: 'utf-8' codec can't decode byte 0xfc in position 38: invalid start byte
This is a notorious issue with python and UTF-8/Unicode support.
Any chance there is "unicode"/utf8 code in the uncommitted changes section ?
BTW you can set an environment variable before spinning the agent, telling it always to use UTF8set PYTHONUTF8=1
Hi GiganticTurtle0
You can keep clearml following the dictionary auto updating the UI
args = task.connect(args)