I have tested the python code on youtube dataset and prone only costs 390 seconds, impressive! But the memory cost is another important factor I concerned and prone-py uses 12G+ memory, I wonder if the cpp code costs less memory and how much it costs on youtube dataset? Thanks!
PS: I'd like test it myself, but install icc compiler and other cpp library is not easy, could you please give me a quick answer, thanks!
I have tested the python code on youtube dataset and prone only costs 390 seconds, impressive! But the memory cost is another important factor I concerned and prone-py uses 12G+ memory, I wonder if the cpp code costs less memory and how much it costs on youtube dataset? Thanks!
PS: I'd like test it myself, but install icc compiler and other cpp library is not easy, could you please give me a quick answer, thanks!