The week started on a wednesday, an hour of Python figuring out some input/output elements. Slightly unsuccessful for the task, but after verifying with some colleagues I was barking up the wrong tree.
Thursday I was unable to spend an hour coding due to work constraints. This was replaced with some reading on the commute to-and-from work involving a quick refresher on basics.
Friday, 1 hour on the Code of Kutulu challenge. 1 Hour was clearly not enough but I think I am now grasping how it works. I should be able to write some functions tomorrow.
Week 2: (97 days to go)
Saturday spend some time working on the Code of Kutulu challenge. Ranked up one league, but finding the closest explroer and moving to that object is not enough to outlast the rest of the AI. Going to need to figure out how to use the given map as a way to find a scaperoute and move away from the incoming wanderers. Attempted refactoring to perl, but not very succesful.
Sunday attempted to refactor the perl code but clearly I need to work on that after a couple years of little practice with the language. Returned to python and the decision algorithm seems to be working. Definitely not pretty though. Need to start using a path analysis algortithm to move away from the wanderers.
Monday refactored the code so that the decision algorithm is in a function rather than hardcoded, this allows to test different strategies without losing either one of them. Added the ability to read the map and find spawn points of the wonderers. Idea is to the use this to send the bot to a different area.
Tuesday corrected the conditions for the explorer to plan. Survived a bit longer, but not by much more. Should probably not be hard coded, but unsure of what to base this on.
Wednesday Added the axis control. If a wanderer is on the same axis as the explorer, explorer will try running in the opposite direction as far as possible.
Thursday, Friday Following a hectic week I took these 2 days to be ‘computer-less’. Started reading up on FLASK and REST for the code jam starting on Monday.
The Code Kututlu challenge was interesting. Given more time I would have invested it into fine-tuning the decider function so that it would take into account more elements, allowing it to possibly better decide in which direction to go.