Dejan Štepec

@po0ya but aren't the weights (*.caffemodel) that are saved by the default solver already normalized (because they were never unnormalizied, because the caffemodel was not saved using provided snapshot functionality)....

soft-nms is used completely independently of the model that has been learned as it only processes/filters detections based on their scores and IoU scores (overlapping measure). There is possibility of...

I think we should talk in terms of epochs instead (the training is printing that). @ZheC when you mentioned that you were using 2 x Titan X with batch size...

I have created LMDB from COCO dataset similar as in the original CPM work bypassing json files (directly creating LMDB from COCO API). I have skipped persons that are smaller...

@jaybdub-nv I can report the same issue. Reboot does not help. Memory consumption is constantly around 6.5GB but in resnet_v2_152 goes through the limit of 8GB.

Thanks! I will take a look at Mongo. What would you suggest, given that I already have deployed DSA and in active use? Can I deploy an annotation tracker to...

@manthey I have forks for DSA, HistomicsUI, and slicer_cli_web_repos due to some minor changes on my side. I have now merged them with upstream to be up to date. What...

I tried it now and for me, it doesn't improve the performance bottlenecks, but hopefully, I have the latest versions. I did: ./deploy.sh ... remove to clean the old images...

Ah, I have now cleaned all the old docker images manually and redeployed the DSA and now the bug seems to be fixed. It's really much more responsive! Next week...

I think that the problem is in Girder Worker Docker container nvidia/GPU support check: https://github.com/girder/girder_worker/blob/master/girder_worker/docker/tasks/__init__.py#L74 https://github.com/girder/girder_worker/blob/e635acedb50d78d48de227bad7cc2bf7376810cf/girder_worker/docker/nvidia.py#L1 It seems like you need to have LABEL com.nvidia.volumes.needed=nvidia_driver present in the image for...