Back in college, my Machine Learning professor turned me on to Project Euler. I didn't initially get too far through the challenges due to other projects, school work etc. taking up most of my time; I think I was doing the challenges mostly in C++ too which isn't the easiest language to start churning these problems out in.
A couple of years ago I decided to pick back up on the Project Euler challenges using Python. I've completed about 30 of the ~600 some-odd challenges, a pretty minimal percentage, I know but you have to start somewhere right?
It's been about a year since I have been regularly working on the challenges. My hope is that by checking my solutions in I'll be motivated to start working through more of the challenges. Also, some of these solutions exhibit extreme brute force mentality and could really be better optimized which is something I'd also like to re-visit.
The original questions aren't included in the solutions but can be found at https://projecteuler.net/archives. It's not hard to find solutions for the challenges either in case you want to fact check me or find a better solution than mine, I'll admit, I've had to visit a couple of these solutions myself to see how others approached solving them.
I've broken down each solution into its own file. Most of the files make use of a timer for benchmarking. Some of the earlier brute force solutions were refactored to be quicker and more efficient, mostly measured by execution time, leading to timers being used in many of the solutions. Each solution outputs the answer which should also be commented in the code as well.
The solutions should be compilable with Python 2.7. The only additional library I've really made use of is Numpy. I don't expect to use any other libraries going forward but I wouldn't rule that out just yet. I'd like to keep my solutions as vanilla as possible, I mean if I'm not writing the algorithms myself does it really count as a "solution"? I personally enjoy using the PyCharm IDE from Jetbrains but of course that's just my personal preference.
- Obviously, solving more of the challenges.
- Optimizing existing solutions.
- Better organization of the source code. Maybe being able to run and output the solution to a challenge given the challenge # from a prompt?
- Implementing the solutions in different languages?