SteadyFox10

10 Questions, 49 Answers

Active since 10 January 2023

Last activity one month ago

Reputation

Badges 1

44 × Eureka!

Questions 10
Answers 49

0 Votes

2 Answers

735 Views

0 Votes 2 Answers 735 Views

Hi Trains Community, I See That For The

Hi trains community, I see that for the trains-agent you have a docker option, did you plan to have any other container system like https://sylabs.io/docs/ i...

clearml

4 years ago

0 Votes

6 Answers

620 Views

0 Votes 6 Answers 620 Views

I Use

I use ignite and their ModelCheckpoint system to be able to resume a training in case of unexpected issue (the electricity shut down last week and our cluste...

clearml

4 years ago

0 Votes

6 Answers

632 Views

0 Votes 6 Answers 632 Views

I Want To Retrieve The Logged Metrics To Be Able To Save The Best Model From My Training. This Is My Step:

I want to retrieve the logged metrics to be able to save the best model from my training. This is my step: trains_logger = Task(...) args = parse_commandline...

clearml

4 years ago

0 Votes

7 Answers

602 Views

0 Votes 7 Answers 602 Views

Is It Possible To Select A Bunch Of Experiment And Archive Them All At Once ? I Tried With The Checkbox But There Is No Option To Archive Them All. I Do It One By One By Hand At The Moment.

Is it possible to select a bunch of experiment and archive them all at once ? I tried with the checkbox but there is no option to archive them all. I do it o...

clearml

3 years ago

0 Votes

9 Answers

630 Views

0 Votes 9 Answers 630 Views

I Use The

I use the docker-compose.yml file and keep everything inside /opt/trains : tree -d . ├── config ├── data │ ├── elastic │ │ └── nodes │ │ └── 0 │ │ └── _state...

clearml

4 years ago

0 Votes

16 Answers

749 Views

0 Votes 16 Answers 749 Views

When Using

when using trains-server is there any limitation for the debug images in Metrics & Logs ? This small script show the behavior, the iteration kept is only the...

clearml

4 years ago

0 Votes

30 Answers

716 Views

0 Votes 30 Answers 716 Views

I Have A Question Regarding The Deletion Of Archived Experiments. Some Of Them Can'T Be Deleted And The Error Message Is

I have a question regarding the deletion of archived experiments. Some of them can't be deleted and the error message is General data error (TransportError(5...

clearml

2 years ago

0 Votes

5 Answers

594 Views

0 Votes 5 Answers 594 Views

I Remember There Is An Environment Variable To Specify Where My

I remember there is an environment variable to specify where my trains.conf | clearml.conf is but I could not remember the name of this env var. Can someone ...

clearml

3 years ago

0 Votes

6 Answers

649 Views

0 Votes 6 Answers 649 Views

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

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...

clearml

3 years ago

0 Votes

1 Answers

663 Views

0 Votes 1 Answers 663 Views

I Think I Have Found An Issue In The Model Visualisation, When You Click On A Model You See The General Tab On The Right, If You Click On Network Or Label And Go Back To The General Tab All The Field Are Now N/A Etc You Could See The Behavior Here:

I think I have found an issue in the model visualisation, when you click on a model you see the general tab on the right, if you click on network or label an...

clearml

3 years ago

0 Hi There, I'M Training A Pytorch Model And Save It Every Epoch. It Seems Like The Model Wights Are Overridden And I Can'T Choose The Best Model After The Experiment Ends. This Feature Is Missing Or I'M Not Using The Library Well?

Well, I use ignite and trains-server with a logging similar to ignite.contrib.handlers so I will be very happy to test this integration.

4 years ago

PompousBeetle71 , check the n_saved parameter on the ModelCheckpoint creation.

4 years ago

0 Is It Possible To Select A Bunch Of Experiment And Archive Them All At Once ? I Tried With The Checkbox But There Is No Option To Archive Them All. I Do It One By One By Hand At The Moment.

now that I know I could use the right click I'll use it like in google drive etc.

3 years ago

0 Is It Possible To Select A Bunch Of Experiment And Archive Them All At Once ? I Tried With The Checkbox But There Is No Option To Archive Them All. I Do It One By One By Hand At The Moment.

Oh right click! Nice, I don't even right click on web page usually, that pretty nice. Thanks.

3 years ago

0 I Use

Wow, that really nice. I did almost the same code, TrainsLogger , TrainsSaver , and all the OutputHandler . I'll use your version instead and put any comment if I find something.

4 years ago

0 Is It Possible To Select A Bunch Of Experiment And Archive Them All At Once ? I Tried With The Checkbox But There Is No Option To Archive Them All. I Do It One By One By Hand At The Moment.

Maybe just adding the use case in the document or elsewhere like video idk could be useful.

3 years ago

0 I Want To Retrieve The Logged Metrics To Be Able To Save The Best Model From My Training. This Is My Step:

It works well I just need to use the task.id for the task='' I tough I could use the task.name it's perfect thanks AgitatedDove14 .

4 years ago

0 I Have A Question Regarding The Deletion Of Archived Experiments. Some Of Them Can'T Be Deleted And The Error Message Is

Is it better on clearml or clearml-server ?

2 years ago

0 When Using

I made the experiment on the allegroai demo server and it's the same https://demoapp.trains.allegro.ai/projects/fcf3f3fb1013434eb2001870990e5e94/experiments/6ed32a2b5a094f2da47e6967bba1ebd0/output/debugImages . I really think it's a technical limitation to not display all the image am I right ?

4 years ago

0 I Want To Retrieve The Logged Metrics To Be Able To Save The Best Model From My Training. This Is My Step:

Wow thanks a lot, I'll test it. I didn't even search in trains_agent documentation.

4 years ago

0 I Want To Retrieve The Logged Metrics To Be Able To Save The Best Model From My Training. This Is My Step:

Is it possible to get all the iteration for one specific metric ? Lets say I have this metric logged. Will I be able to retrieve these series ?

4 years ago

0 Hi Trains Community, I See That For The

I'll give a try with the virtualenv solution. If I have any question I'll ask in this thread. Thanks a lot.

4 years ago

0 I Use

Thansk for the ouput_uri can I put in the ~/trains.conf file ?

4 years ago

0 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

AgitatedDove14 This is what I expected for the community version. It would really nice to have a read-only link. My use case is when I have a merge request for a model modification I need to provide several informations for our Quality Management System one is to show that the experiment is a success and the model has some improvement over the previous iteration. Obviously I don't want the reviewer to see all my failed experiments 😉 . So yes it should be really nice to have read-only ...

3 years ago

0 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

Yes this will work I think.

3 years ago

0 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

Oh can't wait to see this feature 👀

3 years ago

0 Hi, Does 'Trains' Save Scalars Somewhere In The Machine Similar To Log File?

To retrieve metrics from an experiment I use this:
` from trains_agent import APIClient

client = APIClient()

client.events.get_scalar_metric_data(task=task_id, metric="name_of_metric") `Thanks to AgitatedDove14 that point this to me.

3 years ago