Hi DilapidatedDucks58
is this something new ?
usually copy pasting directly from the UI parses everything, no?
CLI? Code ?
Also, How do I make the files other than entry script visible to the job?
The assumption for clearml (regradless on how you create a Task) is that you code is either a standlone script (or jupyter notebook) or inside a git repository. In case of a git repository cleamrl-agent will clone the git repository of the code, apply the uncommitted changes and run your code.
Hi StraightCoral86
When I run an experiment using
Task.create()
,
Use Task.init
🙂
Task.create is meant to create an extranl Task (i.e. Job) ins the system, Not to auto-gernerate a job from the running code. Make sense ?
Ohh StraightCoral86 did you check cleaml-task
? This is exactly what it does
(this is the CLI, from code you basically call Task.create & Task.enqueue)
Will this solve it ?
If a Task is in the 'Completed' I think the only option is to 'Reset' it (see image).
In the UI yes, in code you can do task.mark_aborted(force=True)
You do clear the previous run execution but I think for a repetitive task this is fine.
I would avoid that, no?
Hi BeefyHippopotamus73
. I checked the template task and the list of “Installed Packages” indeed does not have one of my required packages in the list.
Basically the "installed packages" is auto populated based on the directly imported packages n your code base.
Could it be you do not have import snowflake-connector-python
and this is a derivative package (i.e. required from a different package)
BTW: when you clone your Task in the UI you can edit and add the missing packages,...
Hi ReassuredOwl55
How would I find Tasks that have the same code with different inputs/parameters?
Assuming you have the git repo
you can do:Task.query_tasks(..., task_filter={'_all_'=dict(fields=['script.repository'], pattern='github.com/user/repo'))
wdyt?
Hi @<1523701066867150848:profile|JitteryCoyote63>
Hi, how does
agent.enable_git_ask_pass
works
basically it pushes the pass through stdin to git when it asks (it is a git feature)
Check the log, the container has torch 1.13.0 but the task requires torch==1.13.1
Now torch package inside those nvidia prepackaged containers are compiled a bit differently . What I suspect happens is the torch wheel from pytorch is not compatible with this container . Easiest fix , change the task requirments to 1.13
Wdyt ?
Metadata might be expensive, it's a RestAPI call, and we have found users putting hundreds of artifacts, with preview entries ...
That said, you might have accessed the artifacts before any of them were registered
However, are you thinking of including this callbacks features in the new pipelines as well?
Can you see a good use case ? (I mean the infrastructure supports it, but sometimes too many arguments is just confusing, no?!)
PompousParrot44
you can always manually store/load models, example: https://github.com/allegroai/trains/blob/65a4aa7aa90fc867993cf0d5e36c214e6c044270/examples/reporting/model_config.py#L35 Sure, you can patch any frame work with something similar to what we do in xgboost, any such PR will be greatly appreciated! https://github.com/allegroai/trains/blob/master/trains/binding/frameworks/xgboost_bind.py
Guys FYI:params = task.get_parameters_as_dict()
This looks like 'feast' error, could it be a configuration missing?
@<1587253076522176512:profile|HollowPeacock33>
Is this a commercial ad? this seems like out of scope for this channel
Can you expand?
@<1523701304709353472:profile|OddShrimp85> are you trying to shut down the one running on your machine ?
Basically it gives it direct access to the host, this is why it is considered less safe (access on other levels as well, like network)
Since my deps are listed in the dependencies of my setup.py, I don't want clearml to list the dependencies of the current environment
Make sense 🙂
Okay let me check regrading the "." in the venv cache.
Hi @<1562973095227035648:profile|ThoughtfulOctopus83>
The host should be just the host name, no https prefix, I'm assuming that's the issue
Hi ColossalAnt7
Following on SuccessfulKoala55 answer
I saw that there is a config file where you can specify specific users and passwords, but it currently requires
- mount the configuration file (the one holding the user/pass) into the pod from a persistent volume .
I think the k8s way to do this would be to use mounted config maps and secrets.
You can use ConfigMaps to make sure the routing is always correct, then add a load-balancer (a.k.a a fixed IP) for the users a...
Hi WittyOwl57
I think what happens is it auto-logs the joblib load/save calls, these calls track models used/created by the code, and attach them to the model repository representing these model.
I'm assuming there are multiple load/save , and there are multiple model instances pointing to the same local file "file:///tmp/..." . The earning basically says it is re-registering existing models.
Make sense ?
while I'm looking to upload local weights
Oh, so this is not "importing uploaded (exiting) model" but manually creating a Model.
The easiest way to do that is actually to create a Task for Model uploading, because the model itself will be uploaded to unique destination path, and this is built on top of the Task.
Does that make sense ?