Try to set this line in your clearml.conf to true:
https://github.com/allegroai/clearml/blob/6e6271fb91f2aeb2aa7a13c6d07d4e635baaa670/docs/clearml.conf#L177

I mean what is the actual link?
File:// is a path to a file.
If your machine cannot access that path you get an error.
For example:
file:///home/user/file.bin
translates to /home/user/file.bin
If you do not have the file /home/user/file.bin on your machine you get an error.
GrievingTurkey78 make sense ?
Note that by default trains / clearml will not upload your weights file anywhere , only if you set "output_uri" to a specific location it will do that .

So maybe the path is related to the fact I have venv caching on?

hmmm could be...
Can you quickly disable the caching and try ?

Hi SpicyLion54
the -f flag is not very stabe for pip (and cannot be added in requirements.txt). ClearML agent mwill automatically find the correct torch (from the torch repository) based on the cuda it detects in runtime.
This means it automatically translates torch==1.8.1 and will pull form the correct repo based on torch support table.

Sounds good, I assumed that was the case but I was not sure.
Let's make sure that in the clearml.conf we write it in the comment above the use_credentials_chain option, so that when users look for IAM roles configuration they can quick search for it 🙂

Good news, there is an offline mode.
Task.set_offline(True)
If you want your code to be aware, you can do:
from trains import Task if Task.current_task(): Task.current_task().get_logger().report_confusion_matrix(...)

LudicrousParrot69 this is implementation issue, this entire page is based on "task comparison" single Task means totally different interface for querying the data 🙂

Hi SubstantialElk6 I'll start at the end, you can run your code directly on the remote GPU machine 🙂
See clearml-task documentation, on how to create a task from existing code and launch it
https://github.com/allegroai/clearml/blob/master/docs/clearml-task.md

That said, the idea is that you add the Task.init call when you are writing/coding the code itself, then later when you want to run it remotely you already have everything defined in the UI.
Make sense ?

I appended python path with /code/app/flair in my base image and execute

the python path is changing since it installs a new venv into the system.
Let me check what's going on with the pythonpath, because it is definitely is changed when running the code (the code base root folder is added to it). Maybe we need to make sure that if you had PYTHON PATH pre-defined we restore it.

CurvedHedgehog15 is it plots or scalars you are after ?

SubstantialElk6 try to add -e CLEARML_AGENT_EXTRA_PYTHON_PATH=/code/app/flair
It should add it to the runtime pythonpath
(to the BASE DOCKER IMAGE on the Task itself)

JitteryCoyote63
Yes this extremely annoying, I think it was updated on the community server, let me check if we deployed a new docker with a fix ...

Sounds good.
BTW, when the clearml-agent is set to use "conda" as package manager it will automatically install the correct cudatoolkit on any new venv it creates. The cudatoolkit version is picked direcly when "developing" the code, assuming you have conda installed as development environment (basically you can transparently do end-to-end conda, and not worry about CUDA at all)

but this will be invoked before fil-profiler starts generating them

I thought it will flush in the background 😞
You can however configure the profiler to a specific folder, then mount the folder to the host machine:
In the "base docker args" section add -v /host/folder/for/profiler:/inside/container/profile

Here is a nice hack for you:
Task.add_requirements( package_name='carla', package_version="> 0 ; python_version < '2.7' # this hack disables the pip install")This will essentially make sure the agent will skip the installation of the package, but at least you will know it is there.

(Do notice that even though you can spin two agents on the same GPU, the nvidia drivers cannot share allocated GPU memory, so if one Task consumes too much memory the other will not have enough free GPU memory to run)

Basically the same restriction as manually launching two processes using the same GPU

I see... Let me check something

Bad news, there isn't a nice interface to get the table from the Optimizer object (I will make sure we add it, no reason not to).
But you can very easily get all the information you need and more:
all_the_tasks = an_optimizer.get_top_experiments(top_k=100)Then for every task in the list you can get All the information:
for task in all_the_tasks: task_params_as_dict = task.get_parameters() task_scalars = task.get_last_scalar_metrics()Basically the Task object enables you to que...

Yes, I was referring to logging the "clearlm-data" Dataset ID on the Task itself, not an external database.
Make sense?

LOL AlertBlackbird30 had a PR and pulled it 🙂
Major release due next week after that we will put a a roadmap on the main GitHub page.
Anything specific you have in mind ?

I want to use services queue for running services, and I want to do it on k8s

So yes, as a standalone pod with the agent in venv mode (as opposed to docker mode)
Does that make sense to you?

Oh my bad, post 0.17.5 😞
RC will be out soon, in the meantime you can install directly from github:
pip install git+

Hi FiercePenguin76
It seems it fails detecting the notebook server and thinks this is a "script running".
What is exactly your setup?
docker image ?
jupyter-lab version ?
clearml version?
Also are you getting any warning when calling Task.init ?

YEY

I think the limit is a few GB, I'm not sure, I'll have to check
And yes the oldest experiments will be deleted first (with the exception of published experiments, they will be deleted last)

My pleasure

So this is a UI visual issue?

Could you test if this is working:
https://github.com/allegroai/clearml/blob/master/examples/reporting/matplotlib_manual_reporting.py

Sure thing, my pleasure 🙂

LOL I see it now 🙂