What is the suggested way of running trains-agent with SLURM? I was able to do a very naive setup: trains-agent runs a slurm job. It has the disadvantage tha...

Hi 👋 I am trying to set up a trains server on GCP. I followed all the steps listed here https://allegro.ai/docs/deploying_trains/trains_server_gcp/ . I also...

If I am using the demo servers, do I need to do something special to use report_image ?

I've been working a bit with trains-agent, having them deployed on different machines listening to queues (docker mode) and it's been working good so far. My...

Is there an elegant way to download all images posted in “debug_samples” from the trains server? 🙂

Hi! I am trying to download data from GS using StorageManager.get_local_copy() . It works fine when I point it to a file i.e gs://bucket/dataset/image.png bu...

Is there an auto scaling solution for GCP users?

I need to create some meta-analysis of my experiments. Is there a "dashboard" view for trains that I can create plots for all experiments metadata? If not, i...

Question about the file server. Currently, we have a machine with MINIO installed, and all file communication is made using the MINIO SDK client. [MINIO is j...

I wanted to ask about html reporting, if I want to do a more fancy visualization (like overlay of two images depending on mouse hovering), I have to inject t...

I have the slack server running at localhost:8080 When trying to access it from a remote computer, I am getting a screen like so: How can i see the dashboard...

Hi all! Is there a way for trains to recognize the CLI arguments when using https://github.com/google/python-fire instead of argparse?

hey ya'll just joined

who else besides christine got this error?

Hi! I just deploy train-server with all components on Kubernetes and after some modifications, it works. However apiserver and files are failing health-check...

How can I avoid TRAINS Task: overwriting (reusing) the overwriting? I want to launch multiple runs with different hyper parameters on the same machine, yet t...

hey, trying to use trains-agent to run an experiment on my computer. when trying to execute a job from the queue on my agent im getting an error that numpy i...

I'm running multiple trains-agents on one BFG-9000 machine. Is there a way to stop only one agent? trains-agent daemon --stop is my only option right now, bu...

I rebooted my server in AWS, and now the instance is acting weird. The web login is gone, and all my projects are not visible. I tried to use the docker-comp...

Is there any status update on what's going on Train's integration with PyTorch Lightning? What needs to be done before it's usable again? This GitHub issue s...

Hi, the link to the StorageManager docs is broken in the example notebook here: https://github.com/allegroai/trains/blob/master/examples/reporting/Allegro_Tr...

Hi there, I'm integrating trains - looks great so far! We wanted to be able to monitor which user created an experiment. I've noticed I can create a user usi...

Default installation of trains-server exposes only logging via http not https, in case the server has a public IP, what is the recommended way for increasing...

When launching a task to trains agent, I'm having trouble getting the imports from other files working correctly. For instance, if my task imports a function...

Hi! Regarding the artifact.get_local_copy() method, since there is no way to specify the path where the artifact will be downloaded, I wanted to confirm that...

I am also experiencing a weird behaviour when running a script using the module flag. For example I run: python -m module.script arg1 arg 2And after the scri...

Hi, i have configured trains to run in kubernetes. The dashboard tries to access the api by using trains.domain:8008. Can I change this by appying a configur...

Hi all 🙂 is it possible to select two differently named scalar plots in the comparison and put them on the same axis?

Question about the storage manager. Assuming I have an object that updates frequently and always saved at the same path (e.g. my_bucket/my_data.csv ) and I w...

Hello, I am getting ValueError: Could not get access credentials for ' s3://my-bucket ' , check configuration file ~/trains.conf but I did specify them in my...