23 фев 2015 г. - Speeding-up convolutional neural networks using fine-tuned cp-decomposition (ICLR 2014)
Описание:
Four-component Decomposition结果：
For the standard ImageNet architecture (AlexNet), the approach speeds up the second convolution layer by a factor of 4x at the cost of 1% increase of the overall top-5 classification error.
Добавлено на ленту времени:
Дата:
23 фев 2015 г.
Сейчас
~ 10 г назад