This repo includes experiments performed for understanding the speedup convolutions offer by utilizing Gemm
This repo contains experiments from my blog Speeding Up Convolutions that I have performed using native C++.
Libraries such as OpenBLAS and GEMM obviously perform faster than the experiments performed above. This experiments are only conducted for learning purposes and to understand how convolution is made faster for execution using algorithms.
If you have any changes that might improve above algorthms, all ears. Happy Learning :)