Mojo for Evolutionary Algorithms
hey everyone! I'm currently working on a semester project comparing the performance of evolutionary algorithms in Mojo vs python, which I plan to expand into my bachelor's thesis (if I'm able to finish it). However, I'm struggling to achieve the expected performance gains in Mojo due to some challenges with list comprehensions, creating lists of lists, and using random effectively etc.
as a bachelor student, I'm quite unsure whether I should keep trying to figure out how to optimize performance in Mojo or find a new idea before it's too late. I'm curious if it's well suited for evolutionary algorithms and whether the performance gains I'm seeking are achievable without excessive difficulty.
any advice on that matter would help me a lot
2 Replies
Mojo is definitely rough around the edges atm and there are lots of bugs and gotchas I've run into. If you can't find an answer in the docs or via the bot I'd just post a specific question about how something works or how to accomplish something here in the questions channel and someone will help you after not too long.
I also recommend you ask more specific so you can get specific support from the community. Like for example the standard Mojo dictionary is quite slow right now, so if you make extensive use of it, it might be the bottle neck. But without seeing your code its hard to give any adivce