My solutions to the Advent of Code problems.
- Day 1: Trebuchet?!
- Day 2: Cube Conundrum
- Day 3: Gear Ratios
- Day 4: Scratchcards
- Day 5: If You Give A Seed A Fertilizer
- Day 6: Wait For It
- Day 7: Camel Cards
- Day 8: Haunted Wasteland
- Day 9: Mirage Maintenance
- Day 10: Pipe Maze
- Day 11: Cosmic Expansion
- Day 12: Hot Springs
- Day 13: Point of Incidence
- Day 14: Parabolic Reflector Dish
- Day 15: Lens Library
- Day 16: The Floor Will Be Lava
- Day 17: Clumsy Crucible
- Day 18: Lavaduct Lagoon
- Day 19: Aplenty
- Day 20: Pulse Propagation
- Day 21: Step Counter
- Day 22: Sand Slabs
- Day 23: A Long Walk
- Day 24: Never Tell Me The Odds
- Day 25: Snowverload
- Day 4: Don't brute force through all cards on task 2
- Day 5: Find a better solution for task 2 (runs in ~1:29)
- Day 6: Vectorize w/ numpy
- Day 9: Is there a way to remove the need for lists on part 2?
- Day 16: Currently takes 11 cpu hours to run. I have no idea why mine takes so much longer. Could be that I'm copying the entire list for a beam? Moving around all that memory would be annoying. As far as I'm aware, I'm bailing out at the first possible instance.
- Day 17: ... Pathfinding is not my strong suite.
- Day 20: Convince myself the part 2 logic makes sense
- Day 22: Can we drop the bricks faster?
- Day 23: Currently could be simulating each sub_path more than once, cutting back would be useful