Yea, I don't do anything manually.

Quick question: Where again does clearml place the venv? I wanna take a look into it after the task has failed

Seems possible because I didn't know I had to specify an entrypoint somewhere. I will do some additional tests.

Thanks.

With remote_execution it is command="[...]" , but on local it is command='train' like it is supposed to be.

Installs CPU

I tried "~=", "==" and "="

For example in our case we do reinforcement learning and the we would call a script like this: python run_openai_gym.py some_ http://package.my _agent .

Yes. What do you mean by unconnected?

Hey, that is unfortunately not possible as there are multiple projects from different users.

latest for everything.

Thanks for answering. I don't quite get your explanation. You mean if I have 100 experiments and I start up another one (experiment "101"), then experiment "0" logs will get replaced?

Perfect, thank you 🙂

Here it is:

Yea I know, I reported this 🙂 .

Perfect, just what I always wanted. Looking forward to the MinIo version. Thank you:)

Specific step in the pipeline. The main step (the experiment) is currently just a file with a Task.init ` and then the experiment code. I am wondering how to modify this code such that it can be run in the pipeline or as standalone.

Thank you for helping!

Thank you very much for the quick answer. Still so confusing to me that so many things are configured client side 😄

To answer my own question: In the WebUI where one inputs the credentials, use https for the host instead of the auto-added http

` =============
== PyTorch ==

NVIDIA Release 22.03 (build 33569136)
PyTorch Version 1.12.0a0+2c916ef ... Looking in indexes: ,
Requirement already satisfied: pip in /root/.clearml/venvs-builds/3.8/lib/python3.8/site-packages (22.0.4)
2022-04-07 16:40:57
Looking in indexes: ,
Requirement already satisfied: Cython in /opt/conda/lib/python3.8/site-packages (0.29.28)
Looking in indexes: ,
Requirement already satisfied: numpy==1.22.3 in /opt/conda/...

Python 3.8.8, clearml 1.0.2

If you compare the two outputs it put at the top of this thread, the one being the output if executed locally and the other one being the output if executed remotely, it seems like command is different and wrong on remote.

These are the errors I get if I use file_servers without a bucket ( s3://my_minio_instance:9000 )

2022-11-16 17:13:28,852 - clearml.storage - ERROR - Failed creating storage object Reason: Missing key and secret for S3 storage access ( ) 2022-11-16 17:13:28,853 - clearml.metrics - WARNING - Failed uploading to ('NoneType' object has no attribute 'upload_from_stream') 2022-11-16 17:13:28,854 - clearml.storage - ERROR - Failed creating storage object ` Reason: Missing key...

Yea, the one script that is preinstalled.

Restart did not fix it, but somehow looking at tasks works again.

Mhhm, now conda env creation takes forever since it probably resolves conflicts. At least that is what is happening when I tried to manually install my environment

Ah, thank you very much

I created an github issue because the problem with the slow deletion still exists. https://github.com/allegroai/clearml/issues/586#issue-1142916619

Can you explain what you meant by entropy point file? In a new git repository my code works fine.