Hi everyone, I am trying to use docker mode for trains-agent, but it seems that it has problem with the use of multiple gpus this is my trains-agent command:...

Hi everyone, I'm trying to execute trains-agent in docker mode with conda as package manager, is it supported? I tried to work with nvidia/cuda:10.0-runtime-...

Hi, I'm trying the experiment tracking (metrics/plots) and when comparing plots between two experiments, only matplotlib plots (autologged, confusion matrix ...

Hello! We are trying to upgrade from Trains server 15.1 to 16.1 using Docker, but are running into a permission error: trains-elastic | "stacktrace": ["org.e...

Hey there, is there a way to access the trains configuration programmatically at runtime in a task (the configuration that is dumped by the agent in the logs...

Is it possible to view the actual code of a task? As in the script that created the task?

Hi, I'm getting error 404 when trying to see debug samples I've recorded with record_image. the local path I've provided is valid (image is displayed normall...

Could you please explain a bit more how trains adapt the torch version depending on the installed cuda version? Here is my setup: cuda 102 installed and corr...

Hey guys, quick question: is there a tool function to know if a task id is valid? Not verifying that the task itself exists, just that the task id is the cor...

Hey guys, In your opinion, what the best way to upload an artifact to an existing experiment from a storage-server (e.g., S3)? In the storage module document...

Hi, I was getting a really weird error due to mismatch on the versions between the installed libraries in my environment and the ones ran in the node (I manu...

Latex in plot labels? I tried different things, like adding r'{LATEXCODE}' but it does not seem to get recognized. Am I missing something? I am using the ten...

Hey guys, Anyone knows what it means if i deployed a new trains server and when i access it My tab looks like this?

Bug reporting: Just updated to trains==0.16.3. Tested with lightgbm==3.0.0 File "/home/nokchan/anaconda3/envs/cba/lib/python3.7/site-packages/trains/binding/...

Are trains able to discover the docker image and docker command that is used for running the current experiment, or do I need to set it up manually by task.s...

In the experiment list page I can add custom columns (like metrics) which is great, but can I save the view to switch between the different views, or at leas...

Hey, Is there a way to report progress timeline in a Data Processing type task? Use case: I have a distributed data preparation process for which I use such ...

How should I go about making all the artifacts default saving location to a bucket? (destination for automatically saved artifacts, uploaded and registered) ...

I have a local version of trains server (installed on a VM in GCP), one problem I run into was that Task.init() would hang, and in the UI I could see "drafts...

Hi, how can I remove a tag from a task via code in a non-barbaric way?

Hi, I have a worker on a machine using gpus 0,1 and another worker on the same machine using gpus 0,1,2,3,4,5. A worker ran a task on gpus 0,1 but for some r...

AFAIU by default trains logs all tensorboard things, can this be turned off?

PyTorch Lightning question about logging a figure. I have the following code: # turn confusion matrix into a figure (Tensor cannot be logged as a scalar) fig...

Question about out-of-Python dependencies. For example NLTK requires apt-get install sqlite-devel how can we manage dependencies like that on the train agents?

Hi, when a step in a pipeline is aborted, it is marked as gracefully finished (painted in blue) and the other steps that depend on it continue. I believe thi...

Hi all, We’re interested in using trains for a new ML project. This project is an early proof of concept so we’d like to start with the open source version. ...

Quick question, can trains log keras loss values and/or metrics automatically? Or would I have to attach a tensorboard callback?

Hey guys, another question about deploying my own trains server. I have a trains-server deployed on my k8s cluster using the trains helm chart (which is awes...

Hi guys, I configured a trains server and a trains agent. I have some code I want to run in the trains agent, however the code is in a local branch on my cli...

Hi everyone, if I want to run a script that has Trains tracking statements in it but just this time I want to disable all logging, how would I go about that?