TensorFlow
Infrastructure
cloud gpu deployment
SHOW HN2015
How HN took it
0 positive, 2 neutral or asking, 2 mixed, 5 critical, out of 9 comments about the project. Thread score -0.38.
Execution: 0 for, 2 againstDocs: 0 for, 2 againstPerformance: 0 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Maker responsive to feedback and updated the README and build
From 2 comments
Thanks for the feedback! I've added a note to the README that support is still experimental. I'll be tracking the issue and updating the repo + AMI as it develops. Will be…
alexkern on HNUpdated to the latest commit, works for me! :)
alexkern on HN
Objections
CUDA 3.0 support causes NaNs and unreliable training
From 2 comments
Honestly, there are enough issues with TensorFlow right now due to CUDA 3.0 that using it with AWS is highly problematic. I appreciate the author's attempt, but there's no way…
Smerity on HNThanks, would be good to have multiple sources of verification that HEAD now supports this natively without issues like unexpected NaNs.…
vrv on HN
Recipe failed: AMI was region-specific and the link was broken
From 3 comments
Top link points to a non-existent AMI so this "recipe" -- like almost every "here's how to ..." recipe on GitHub that gets posted on HN -- doesn't work. I swear, it's like 100%…
tacos on HNI tried with the Oregon region and it failed. I changed to the N. California region and it worked. (No comment about the GPU though.)
droque on HNAn AMI id is region-specific, the original poster should probably have mentioned what region he created the AMI in.
fred256 on HN
Multi-GPU g2.8xlarge not faster than g2.2xlarge
From 1 comment
I think the nan issue you are referring to was caused by some weird stuff with Bazel. I put together an AMI in virginia: ami-cf5028a5 and if you have a masochistic streak, here…
erikbern on HN
Google docs list GPU as unsupported
From 1 comment
Top link points to a non-existent AMI so this "recipe" -- like almost every "here's how to ..." recipe on GitHub that gets posted on HN -- doesn't work. I swear, it's like 100%…
tacos on HN
Requests
Verification that HEAD supports CUDA without NaNs
From 1 comment
Thanks, would be good to have multiple sources of verification that HEAD now supports this natively without issues like unexpected NaNs.…
vrv on HN
Questions
Why multi-GPU instances do not scale under Xen virtualization
From 1 comment
Xen virtualization disables P2P copies ergo GPUs have what we call a "failure to communicate and some GPUs you just can't reach (without going through the CPU that is)."
varelse on HN
Alternatives named
Running TensorFlow locally on a laptop or phone
From 1 comment
Honestly, there are enough issues with TensorFlow right now due to CUDA 3.0 that using it with AWS is highly problematic. I appreciate the author's attempt, but there's no way…
Smerity on HN