Reputation
Badges 1
131 × Eureka!we run in containers without venv, in the main section, and then delete it or use it for similar experiments Sounds like something very similar, I'll try to use it, thanks a lot! Can this be configured in the UI by simply adding a docker image to the launch options?
docker run --rm -v /srv:/root/srv -v /srv/data/apatshin_docker/airflow/dags/voyager:/voyager voyager:cpu python3.7 /voyager/src/pipeline/train_task_dqn_demo.py
For reference
I have now locally raised Clearml on a nearby machine and logging is configured as in the textbook - all metrics are worked out.
On the machine where I run docker, I just copied the clearml.conf file. Maybe you need to do something else (send the /opt/clearml folder and drop it into the docker image?)?
and dell tasks, and all ok:
` #task = Task.init(
project_name=f"RL_experiments/{cfg.train.env_train.target.split('.')[-1]}/{'/'.join(cfg.train.trainmodule.target.split('.')[-2:])}",
task_name="demo",
reuse_last_task_id=False)
#task.connect(dict(OmegaConf.to_container(cfg, resolve=True)))
logger = get_logger("train_ql", log_level=cfg.base.log_level)
logger.info(f"cfg:\n{OmegaConf.to_yaml(cfg)}")
tmp_values = train_dqn_task(cfg.train, cfg.base)
...
TimelyPenguin76
As I remember, the last one is 1.1.4
What exactly do you mean about task logger?
I replaced the open ports with ??9?
so that there would be no conflicts. If you write them down in a new form, then everything is ok
And why can it be that the displayed time is zero?
although the experiments were considered for several days
Docker always has port 8091 File Store Host
- File Store Host configured to: http://localhost:8091
` (base) user@s130:~$ clearml-init
ClearML SDK setup process
Please create new clearml credentials through the profile page in your clearml-server
web app (e.g. )
Or create a free account at
In the profile page, press "Create new credentials", then press "Copy to clipboard".
Paste copied configuration here:
api {
web_server: ` `
api_server: ` `
credentials {
"access_key" = "OV676692R7V...
File Store Host configured to: http://localhost:8090if i setFile Store Host configured to:
then
` (base) user@s130:~$ python3.7
Python 3.7.9 (default, Aug 31 2020, 12:42:55)
[GCC 7.3.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
from clearml import Task
task = Task.init(project_name="my project", task_name="my task")
ClearML Task: overwriting (reusing) task id=bf47e430826d43998c0f54c73addc12b
2021-11-03 19:15:13,491 - clear...
` (base) user@s130:~$ clearml-init
ClearML SDK setup process
Please create new clearml credentials through the profile page in your clearml-server
web app (e.g. )
Or create a free account at
In the profile page, press "Create new credentials", then press "Copy to clipboard".
Paste copied configuration here:
api {
web_server: ` `
api_server: ` `
credentials {
"access_key" = "6VUTS73D48DMPVI1NMPS"
"secret_key" = "4fGtZVirW0ztSLbm6JPLESM...
I didn’t understand how exactly (
in any case, after looking at many examples, I found how it was implemented.
'status_changed': ['>{}'.format(datetime.utcfromtimestamp(previous_timestamp)), ],
we just started logging metrics in clearml
Perhaps this will help
this is the experiment that was useful, but we stopped it, because. convergence has happened before. than we expected
thanks a lot!
Can you please tell me if you know whether it is necessary to rewrite the Docker compose file?
AgitatedDove14
cool
in theory, a calm launch is possible at 1.17.1-2?
@<1523701087100473344:profile|SuccessfulKoala55>
I'm talking about something like OPTUNA
wow, that's interesting, please let me know. Are there screenshots or a demo video somewhere where you can see how the enumeration parameters are set.
i use the local free version of clearml
Is that what the thing was called a bucket?
please tell me, is it possible to somehow make it so that costomous fakets, which are not in the public domain, would be used?
for example, if I somehow start the execution of an agent task in a specific docker container?)
-
specify container from UI
-
libraries in the ubuntu repository have not yet reached their pip / pypi repository
Thank you very much for your help and for such a convenient product!)
I haven't figured out the alents yet, but it already looks amazing!)
I figured it out, I had to click on the model to push, and only after that it would appear as available.
I'm interested in more details)))
please tell me how I can help you?)