Hi, I am new here. I established a successful trains server and everything runs ok. After I changed the ip address of the server I couldn't make it to work a...

Question, Lets say I'm Kaggling, as you might know some "Code Competitions" are restricted from internet access. In the general case, you might have some inf...

Hi allegroal team, I recently saw your project and tried it. This is a great tool! I have a question to ask: How to upload tasks from local machine to server...

Can we somehow add a column "Duration" to the list of Experiments ?

Hi, I have a small issue about GPU monitoring. I run my training inside a Singularity container and I set the CUDA_VISIBLE_DEVICES variable. However, I get t...

Hello! I haven't used trains before, I am looking for opinion from anyone with more experience on whether trains is the correct tool for my non-ML use case. ...

Hi everyone, thx so much for this awesome tool! I was wondering, is there a way to define for trains, which variable in the project is the kpi, and then clus...

Hi my friend mailto:mishael@xfind.ai has gotten his Slack account accidentally deactivated for this workspace. Anyone here can reactivate his account?

Hello! I found today allegro-ai and while running for the first time the example script, I faced this error: Failed auto-detecting task repository: [WinError...

Hi, I've recently upgraded to 0.15.1 from 0.14.2, and for some reason a code that previously worked in which I'm getting the tags of a model using InputModel...

cloning: origin Repository cloning failed: 'NoneType' object has no attribute 'startswith' trains_agent: ERROR: Failed cloning repository. 1) Make sure you p...

pytorch-lightning-bols.loggers.TrainsLogger creates new ids even if reuse_last_task_id=True is set. How can I force it to reuse last ids?

Anyone using trains with Snakemake? I am running my workflow with Snakemake in a docker container, and it can output to the trains server of course, but exec...

Is trains adaptable for federated learning scenarios?

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...

Having issues running trains-server on Win10. trains-elastic exited with code 137 trains-mongo exited with code 100 trains-apiserver exited with code 1 Some ...

Anyone deployed trains on azure, I am interested to know about your experience.

Hi there Trains riders, Is there a built-in way to send notifications upon completed/failed experiment? I have seen the Slack_alerts code sample, where the m...

Hi, I have a train server set up on some IP and accessing the web interface works great. But how do I configure trains (from pip install trains ) to use my s...

Hi, I am trying to execeute my code on nvidia/cuda docker, but it keeps running, it is not failed or not aborted. The last log message is "Successfully insta...

Hi, when it first asks me to enter my full name, it fails to perform the request (timed out). Checked server side and receiving this error [urllib3.connectio...

Hi, first time installing trains-server. But I’m getting the following error when loading the web page Please enable JavaScript to continue using this applic...

Hi Trains team: question - I am trying to implement unbiasing to my datasets. I was wondering if Trains has anything in its toolset that already implements s...

How can I download the plots from 'scalars' and 'plots' in high resolution?

(found it was mentioned in an issue on github)

I'll implement LearnerTensorboardWriter and see the results

Is there a way to get tasks by hyperparameters values? when I use the search in the UI I get the relevant task, but when I try the following I get an empty l...

is there anything I missed in the steps?

Hi guys, I'm new here 🙂 I just launched an AWS EC2 t3.large instance with the AMI from the marketplace according to the docs, and the stage of entering the ...

Hey! I was wondering is it possible to change the ports of the Trains server? We hope to run a trains server locally on our own server, but we already use so...