Re:Ccv open sources near state-of-the-art image classifier u...
by kastnerkyle on Thursday March 27, 2014 @06:01PM
(#7483833)
So... duh. You would need > 1.2 TB to fit all of ImageNet on the card :). Thanks for clarifying, and pardon my brain lapse! Also, thanks for putting this out there - if I get some time I may send some pull requests your way. Awesome stuff.
Re:Ccv open sources near state-of-the-art image classifier u...
by liuliu on Thursday March 27, 2014 @06:24PM
(#7483938)
You need 1.2TB SSD for sure to train the complete ImageNet dataset. The data is loaded into GPU memory only one batch (256 images) at a time. But the loading part will be the bottleneck if you use a rotational disk.
The code is never optimized to the extreme. I optimized the code to the point of being able to finish at reasonable time (9 days for 100 epochs). The convolutional kernel is parametrized (with template and some macro tricks, forward and backward propagation convolutional kernels are parametrized into hundreds of small functions) and the best parameters were chose with a mini benchmark in the beginning of training process.