The implementations
I will post all scripts for this test and I've added a GIT repository, so feel free to test it on your machine. I am also happy if you post some of your solutions with running times :-) I am quite sure that my Java and C++ code can be written much better. If you know how, please leave a comment. If you know other languages, you could create a script for these. I focus on Python, Java and C++ as they are very often used.
I have implemented these three types of algorithms for this post:
- ijk-algorithm: This is a simple, straight forward implementation of a matrix multiplication. I've used the definition of matrix multiplication. I didn't use multiple threads.
- ikj-algorithm: just like the ijk-algorithm, but I've switched two of the three the for-loops.
- Library-functions: I always prefer libraries over self-implemented solutions. I think they are faster than anything I could come up with in a reasonable amount of time.
Read full article from Part I: Performance of Matrix multiplication in Python, Java and C++
No comments:
Post a Comment