Hello, I'm trying to save a keras model as a task artifact, and then upload it from another task. Does anyone know the syntax for that? What I've seen is not...

Hi, we have clearml on k8 setup. Using the below, we run dynamic pods on the cluster. k8s_glue_example.py --queue glue_qI pushed this code, https://github.co...

Hi, Another question There is a way to know if a job is running locally or remotely? Like execute_remotely knows ... > Note If the > task > is running remote...

Hi I came across some inconsistency in the iteration reporting in the ClearML with pytorch-lightning when calling trainer.fit multiple times, before I dive i...

I change agent.default_docker.image in config but when I run job in UI I see that attached old image, why? what I forgot to do?

env variables for clear ml stayed branded as TRAINS? such as TRAINS_HOST_IP not CLEARML_HOST_IP?

apart from the server migration doc , is there another upgrade guide somewhere from trains to clearml?

Hello, I tried the clearml-session CLI to start a jupyter instance on an agent, but an error with the password, here is the full CLI log: $ clearml-session -...

What does it mean to publish a model or a dataset?

Hi, Is there a way to compare the scripts of different experiment runs? This will make it easy to track changes and version control

I'm getting this when running with Keras framework. clearml.storage - ERROR - Failed uploading: [Errno 21] Is a directory: 'model.savedmodel'.

Hi! I was wondering regarding this issue: https://github.com/allegroai/clearml/issues/181 The solution seems to be to not log by default, and only use a cont...

Hi all, I think their is a UI bug. When trying to add experiment to compare:

Hi! I have a GPU workstation at the office (no public ip) with latest clearml-agent installed. When I was in the same network - I was able to use clearml-ses...

Hi! I have a question regarding performances of the clearml-server: are the calls from the agents made asynchronously/in a non blocking separate thread? is t...

Hi! Is there something happening with the ModelCheckpoint callback on tensorflow==2.4.0 ? Using 2.2.0 gave me an input model on the artifacts tab in the GUI 😢

Hi all 🙂 I have a question regarding task.connect_configuration() . Does it possible to update the file on the server when running remotely or locally? for ...

whats different between --cpu-only and --services-mode?

Hey! I'm having a weird issue when I run pip freeze locally it's showing version "clearml==0.17.5rc6" but when I initiate the task it's always starting with ...

Hi, I have a script running cross validation, basically it calls 5 times (5 folds) another script that does a training and evaluation. Is it possible in Clea...

Announcing ClearML 0.17.5 Features Improve Datasets support Reporting and Plotly improvements Add force_download argument (special shout out to JitteryCoyote...

Hi, I'm using the k8s glue and have a few questions. Noted that it's not requesting the http://nvidia.com/gpu thus the pod created doesn't have a GPU resourc...

Btw: There seems to be no support for videos in tensorboard/experiment view (e.g. https://tensorboardx.readthedocs.io/en/latest/tensorboard.html#tensorboardX...

Hello, I have an error while installing git dependencies of local package: So far I used task. update _requirements(“[.]“) with my local package referencing ...

Another strange behavior of the python SDK CLI: after executing python my_task.py, where my_task.py creates and send to the queue an experiment, the command ...

Hi all! Please tell me there are examples of ClearML and pytorch-lightning integration

Hi, I have a self-hosted instance running quite well, pretty good job. I'm wondering if there is any way to have a read-only user? Is it available in the api...

As someone much smarter than me noted.. in our small community we have reached 500 users this morning. It's a small milestone but nothing wrong with celebrat...

Hi friends! I'm trying to upgrade the https://aws.amazon.com/marketplace/pp/B085D8W5NM AMI over to ClearML. The steps seem easy enough, just docker compose d...

Hey all. I'm seeing a strange error when trying to run hyperparameter optimisation by cloning a base training task Action failed <500/0: tasks.clone/v1.0 (ke...