Thanks 😅

AgitatedDove14
It was installed by 'pip install kwcoco' while my conda env was active.
Not sure if it answers your question..

Something like:

model = SomePytorchModel()
checkpoint = {'model_state_dict': model.state_dict()}
torch.save(checkpoint, “model.tar”)

@<1523701070390366208:profile|CostlyOstrich36>
Well, I doubt that it is the case in my situation.
Is there any API where I can get read the metrics per experiment?

No. But this isn't the only package that was installd using pip inside a conda env.
Compare it with Pandas as an example - was installed using pip nad it's actually instaled inside the worker as well (using pip).

what does that actually mean?
2022-07-17 07:59:40,339 - clearml.storage - ERROR - Failed uploading: Parameter validation failed: Invalid type for parameter ContentType, value: None, type: <class 'NoneType'>, valid types: <class 'str'>

Oh wow AgitatedDove14 . Appreciate it!
Are you sure it’s just a matter of the python version?
The same experiment script, was working on the exact docker image in the past (with older clearml versions though…).
For example this experiment log:

Thank for the great explanation! Now it makes much more sense.
You are right about the issue that 'kwcoco' isn't being detected, and Im actually running this as a single script, and the kwcoco not imported directly (but from within another package).
Ill try running it from a repo and see how it works.

Thanks @<1523701070390366208:profile|CostlyOstrich36> .
What are considered as experiment objects?

SuccessfulKoala55
we are using the fileserver which is configured at clearml.conf to a path on a network drive (i.e the NAS) -
files_server: file:///mnt/clearml_storage

SuccessfulKoala55
Is it a rare use case to use a NAS as the fileserver?
What would you suggest?

Its mounted automatically.
The local .ssh folder is copied to a temp folder which is mounted by the agent in the docker run command -
Executing: ['docker', 'run', '-t', ......., '-v', '/tmp/clearml_agent.ssh.fuu4r8ta:/root/.ssh', .... , '--rm', 'nvidia/cuda:10.1-cudnn7-runtime-ubuntu18.04', ....]

AgitatedDove14 , here's the log

AgitatedDove14 its running inside a docker based worker.
Are you interested in the full pip freeze of that docker?

Thanks

The script is intended to be executed remotely.
Can I declare an absolute path in this case?

That would be a very useful feature.
What is the status of that issue? I havn't found it on github.

Absolute in my hard drive?

@<1523701070390366208:profile|CostlyOstrich36> am I doing anything wrong here?

Ok. Thanks!

can you test what happens if you pass the credentials in the global scope as well, i.e. here:

That didn’t help

I did found this:

from clearml.backend_api.session.client import APIClient
client = APIClient()
client.events.get_task_metrics(tasks=['some_task_id'], event_type='log')

but it returns an empty 'metrics' object

Oh!
That was so silly on my side...

Maybe I should understand better how this actually works.
It's not a rare use case to use conda environments, but install some packages using pip (when they're not available at any conda channel). Will that use case be supported on clearml?

TimelyPenguin76

SuccessfulKoala55
I'm not sure I understand your suggestion

Im afraid there's something with my account, since it won't login on other browser as well, but logging with a different email works

Thanks AgitatedDove14 !
I’ll use clearml 1.4.1 until the fix is out.

AgitatedDove14 ,
From the experiment’s console log:
` - boto3==1.16.2

• botocore==1.19.2 `

Not sure I understand the purpose of this.
it meant pip will look for wheels at thus url?