Lab 2: Tensorboard must be run from a GPU node

[willprice]

The login node the students are using don't have a working version of tensorflow installed so they need to run tensorboard in the background instead.

[willprice]

$ python cifar_hyperparam.py &
$ tensorboard --logdir=logs/ --port=<ipnport>

[willprice]

We should request a new module with tensorflow built for CPU usage as the current version only runs on GPU nodes and hence we can't run tensorboard

[dimadamen]

Is the issue related to the GPUs being busy? I still need an hour before I can get to you. Let's see how things progress

…

-- Dima Damen Senior Lecturer in Computer Vision Department of Computer Science University of Bristol, BS8 1UB, UK Tel: +44 117 9545633 http://www.cs.bris.ac.uk/~damen

________________________________ From: Will Price <notifications@github.com> Sent: Monday, October 30, 2017 12:06:44 PM To: COMSM0018-Applied-Deep-Learning/labsheets Cc: Subscribed Subject: Re: [COMSM0018-Applied-Deep-Learning/labsheets] Lab 2: Tensorboard must be run from a GPU node (#7) We should request a new module with tensorflow built for CPU usage as the current version only runs on GPU nodes and hence we can't run tensorboard — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub<COMSM0045-Applied-Deep-Learning/labsheets#7 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AIxOWqEc2lg6iJH9-fZpICnxv_6NeZxlks5sxbvUgaJpZM4QLEoP>.