PokemonGoers / PredictPokemon-2

In this project we will apply machine learning to establish the TLN (Time, Location and Name - that is where pokemons will appear, at what date and time, and which Pokemon will it be) prediction in Pokemon Go.
Apache License 2.0
9 stars 3 forks source link

Add time features #26

Closed MatthiasBaur closed 8 years ago

MatthiasBaur commented 8 years ago

This is in compliance with the new feature system.

jonas-he commented 8 years ago

@MatthiasBaur see #8 . I was also using tzwhere module but found that it caused some bugs with our project (npm install failed). Also tzwhere.init() takes 5 seconds on my system. So I decided to go with an approximation which divides the world into timezones every 15° of longitude.

MatthiasBaur commented 8 years ago

@jonas-he I had the same 5 seconds startup. Really annoying, your solution sounds good.

bensLine commented 8 years ago

@MatthiasBaur good job! we can also add new Date(_pokeEntry.appearedLocalTime) to the pokeEntry directly as time and location are necessary for the calculation of many features. right now we create the date object already quite often.

And thanks for the typo fixes :D