kootenpv/whereami

Uses WiFi signals :signal_strength: and machine learning to predict where you are

whereami

Uses WiFi signals and machine learning (sklearn’s RandomForest) to predict where you are. Even works for small distances like 2-10 meters.

Your computer will known whether you are on Couch #1 or Couch #2.

Cross-platform

Works on OSX, Windows, Linux (tested on Ubuntu/Arch Linux).

The package access_points was created in the process to allow scanning wifi in a cross platform manner. Using access_points at command-line will allow you to scan wifi yourself and get JSON output. whereami builds on top of it.

Installation

pip install whereami

Usage

# in your bedroom, takes a sample
whereami learn -l bedroom

# in your kitchen, takes a sample
whereami learn -l kitchen

# get a list of already learned locations
whereami locations

# cross-validated accuracy on historic data
whereami crossval
# 0.99319

# use in other applications, e.g. by piping the most likely answer:
whereami predict | say
# Computer Voice says: "bedroom"

# probabilities per class
whereami predict_proba
# {"bedroom": 0.99, "kitchen": 0.01}

If you want to delete some of the last lines, or the data in general, visit your $USER/.whereami folder.

Python

Any of the functionality is available in python as well. Generally speaking, commands can be imported:

from whereami import learn
from whereami import get_pipeline
from whereami import predict, predict_proba, crossval, locations

Accuracy

k Generally it should work really well. I’ve been able to learn using only 7 access points at home (test using access_points -n). At organizations you might see 70+.

Distance: anything around ~10 meters or more should get >99% accuracy.

If you’re adventurous and you want to learn to distinguish between couch #1 and couch #2 (i.e. 2 meters apart), it is the most robust when you switch locations and train in turn. E.g. first in Spot A, then in Spot B then start again with A. Doing this in spot A, then spot B and then immediately using “predict” will yield spot B as an answer usually. No worries, the effect of this temporal overfitting disappears over time. And, in fact, this is only a real concern for the very short distances. Just take a sample after some time in both locations and it should become very robust.

Height: Surprisingly, vertical difference in location is typically even more distinct than horizontal differences.

Related Projects

• The wherearehue project can be used to toggle Hue light bulbs based on the learned locations.

Almost entirely “copied” from:

https://github.com/schollz/find

That project used to be in Python, but is now written in Go. whereami is in Python with lessons learned implemented.

Tests

It’s possible to locally run tests for python 2.7, 3.4 and 3.5 using tox.

git clone https://github.com/kootenpv/whereami
cd whereami
python setup.py install
tox