1 point by pythagoras_programmer 1 year ago flag hide 11 comments
johnsmith 1 year ago next
Great work! I've been looking for a lib like this to speed up my GA research. Looking forward to trying it out.
janesmith 1 year ago next
I like how this lib supports multiple mutation and crossover operators. Have you considered adding more docs on how to customize those operators?
nerdherder 1 year ago next
I agree, more docs and examples on how to customize the operators would be very helpful.
andrew 1 year ago prev next
I noticed the lib uses JAX and Numba for acceleration, have you thought about adding support for other platforms? Also, have you noticed any limitations with these libraries?
jax_dev 1 year ago next
JAX is quite flexible and has a lot of nice features. I haven't noticed any limitations with it yet, but let me know if you've had any issues!
codingfanatic 1 year ago prev next
Very cool, looking forward to testing it out! Do you have any benchmark results? Also, how does it compare to other GA libs in terms of performance and features?
optimysm 1 year ago next
In terms of benchmarks, I've noticed a 3x speedup compared to a pure Python implementation. This lib has most of the common GA features, and I've tried to make it as modular as possible for easy customization.
devanalyst 1 year ago prev next
The documentation was very helpful! One suggestion though, it could be more clear on how to use the library with various fitness functions.
codeguru 1 year ago prev next
I've heard using NEAT can sometimes result in better performance than GA. Have you tried testing this lib against others GA or NEAT libs?
aiexpert 1 year ago prev next
I've recently implemented a GA lib myself and found that it's not as simple as just running the algorithm. There are many considerations for implementation that vary depending on the problem. Have your tests taken these factors into account?
learning_lover 1 year ago prev next
I'm new to this topic and have been exploring GA and other optimization techniques. Would you be willing to share how you got started with these, and some tips for someone new in this area?