search

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

Extract features #23

Closed MatthiasBaur closed 8 years ago

MatthiasBaur commented 8 years ago

extractFeatures.js can be easily expanded with more features

bensLine commented 8 years ago

@MatthiasBaur this looks great! Slowly we getting more features :) However, it would be great if you could transform your file into several feature sources, check the readme on https://github.com/PokemonGoers/PredictPokemon-2/pull/25 if it's not merged when you get back.

This allows us to handle features more modular and to integrate them easily into a single file. I guess you would end up with a time and address feature source. longitude and latitude are already provided by the raw data as feature (see API feature source, api_features.js).

MatthiasBaur commented 8 years ago

The changes in this branch where moved into the addTimeFeatures branch

bensLine commented 8 years ago

@MatthiasBaur Ok, if you do not need this branch then please delete it from the repo.