search

accosmin / nano

C++ library [machine learning & numerical optimization] - superseeded by libnano
MIT License
1 stars 0 forks source link

Auto-tunable convolution layer #5

Closed accosmin closed 9 years ago

accosmin commented 9 years ago

When resizing the convolution layer, we should also select the fastest 2D operator (convolution & correlation) automatically using a single thread.