Teach Python how to play ToonTown by utilizing object-orient programming (OOP), design patterns, machine learning (ML), and computer vision. Ultimately, I want to learn how to properly implement and optimize machine learning algorithms when given large amounts of data while also maintaining best practices in Python, OOP, and design patterns.
OmniToon(s) wandering around ToonTown supporting other players (or other OmniToons) in...
- Training Gags
- Completing tasks
- Conquering buildings
- Select random Gag
- Choose random target Cog(s)
- Don't use Sound on Lured Cogs (unless maximizing rewards while training Sound or Lure)
- Heal other Toons if they're low Laff Points
- Move the OmniToon
- Classify Cogs
- Start a Battle
- Read the Gag menu
- Choose Gag and target Cog(s)
- Pass/Run from Battle
- Re-supply Gags from the Gag Shop
- Locate Treasures (health packs) in Playgrounds
- Switch servers if there's a desired/unwanted invasion
- Parse Battle data (Cogs, Toons, attacks, Gags used, Gags available, etc.)
- Collect large amounts of data by running Battle simulations
- Feed data into ML algorithms
- Determine most optimal Gags to use when running various strategies
- Catch fish, turn them in, get Jellybeans for more Gags
- Maximize total Battle rewards
- Watch invasion tracker and switch servers
- Defeat Cogs ASAP (for tasks)
- Learn to stack Gags (everyone uses Lure, Throw, Squirt, etc.)
- Train specific Gags
- Learn Trap/Lure combos
- Don't use Sound on Lured Cogs
- Support other Toons (prioritize Toon-Up, Trap, Lure, etc)
- Optimize movement in order to complete tasks as fast as possible
- Learn ToonTown's neighborhoods, playgrounds, and streets
- Select tasks that are most similar in order to knock out multiple tasks at once