Hi @<1523701070390366208:profile|CostlyOstrich36>
FYI: I’m not using docker agent.

I set the env vars as you mentioned above in config. And I can see the those while starting agent using command: clearml-agent daemon --queue default

While running below code locally eg: python test.py I’m loading the env vars that is working fine.
But my question is when we clone the same task from clearml UI and send it to default queue; So can below piece of code can use those envs?
I tried sendi...

Yes, this is what I’m doing currently. But strugginling to manage the versions and all.

Thanks a lot @<1523701435869433856:profile|SmugDolphin23>
will go through this one.
your quick reply really means a lot. Thanks again!!

Thanks a lot @<1523701070390366208:profile|CostlyOstrich36> 🙏

Thanks @<1523701087100473344:profile|SuccessfulKoala55>

And what I feel; pulling data by using some client eg: minio-client for minio is faster than clearml

sure

it is picking http, it should https?

host: None

Everytime I start training; first download dataset using minio client and then do further operations.

Can we integrate minio easily with clearML?

Well I can understand well your problem. @<1590514584836378624:profile|AmiableSeaturtle81>
Please let me know if you have answers for my problem. 😄

botocore.exceptions.EndpointConnectionError: Could not connect to the endpoint URL: “ None ”

For below config and command it worked
command: clearml-data create --storage " None " --project dummyproject --name testingds2

@<1590514584836378624:profile|AmiableSeaturtle81>
Yes, I can also feel that slowness.

But what about if we have millions of pdfs and we need to generate a image dataset along with their bounding boxes. We cannot generate it on the fly while training the model as it will very very very slow.
So I’m doing all preprocessing the upload preprocessed data back to storage.
And while training again pulling preprocessed dataset from storage. 😄

Hi @<1523701435869433856:profile|SmugDolphin23>
This is follow up question from my side.

Is it efficient to use clearML data for hundred of thousands of image dataset. I’ve a concern about performance.

So basically I’m training a transformers classifier. which need at least 300-400K images.

Though I’m able to connect using minio client with the same creds without any issues.

Hi @<1523701070390366208:profile|CostlyOstrich36>
After removing the port number from command given by you.
Re: your message
with the combination of None :port/bucket for --storage ?

Hi John
Well, I didn’t setup anything extra 🙂

Removed port and set secure as true.

Also tried changing files_server in config but didn’t work

Yup, tried that too.

raise ValueError(“Insufficient permissions (delete failed) for {}“.format(base_url))
ValueError: Insufficient permissions (delete failed) for None

Again throwing Insufficient permissions.

clearml-data create --storage "

" --project dummyproject --name testingds1

clearml-data - Dataset Management & Versioning CLI
Creating a new dataset:

Error: Insufficient permissions (write failed) for ostore.xyz.tech

Am I missing anything in config?

Yes, I tried with default port of minio (9000). didn’t work either.

FYI: I also changed the files_server in config

still got same error

yes