We love seeing all of the social media posts from developers – here are a few highlights from the week:
We run deep learning image recognition to detect 1000 categories on 1 million images – took less than 2 hours on nvidia GPU.
— manovich (@manovich) August 29, 2015
I've updated my website with some CUDA kernels from progressing through the Udacity course ran by NVIDIA. http://t.co/s4HdqZX9P8
— Brady Ledger (@hejbrady) September 1, 2015
Just finished the phenomenal Parallel Programming @nvidia CUDA @udacity course by @jowens + @davedotluebke! My notes: http://t.co/l3qtybGBoH
— Ilya Kavalerov (@IlyaKava) September 1, 2015
Mocha.jl: Deep Learning for Julia http://t.co/PcYqPNby6h I might just give this a try; Torch/lua is quickly becoming too mainstream
— Andrej Karpathy (@karpathy) September 2, 2015
@GPUComputing The power of cuDNN is all around us, stay tuned ;) #nvidia #cuda #cudnn #neuralnetwork
— Alex Z. (@alexzitti) August 29, 2015
@alexjc @doppioslash but after a month of banging my head on the wall I had to buy NVIDIA to work with DeepLearning. cuDNN is blazing fast
— Umar Nizamani (@Rapchik) September 2, 2015
The biggest problem with Xeon Phi is that the average programmer can't have access to it,in contrast to CUDA where a teenager can code on it
— Alexander (Alexandros) Agathos (@agath_alex) September 4, 2015
On Twitter? Follow @GPUComputing and @mention us and/or use hashtags so we’re able to keep track of what you’re up to: #CUDA, #cuDNN, #OpenACC
