Rocero: Auswertung des Füllverhaltens der Glassammelstellen

Wir haben Korrelation zwischen Bevölkerung und Füllverhalten analysiert, sowie eine zeitliche Vorhersage der Füllhöhe pro Sensor errechnet.

Demo

Open Data Hack SG 2023: Füllverhalten von Glas-Sammelstellen

  • Untersuchung der Fülldaten
  • Bereinigung der Fülldaten
    • Ausreisser entfernen
    • Glätten
  • Splitten der Daten in Perioden zwischen Entleerungen
  • Trainieren eines linearen Modells mit linearer Regression

  • Graphen von den Trainieren Modellen

Aufbau des Repos

  • Sensordaten und lineares Modell
    • src/prepare_and_split_data.py: Script um Sensordaten zu Bereinigen
    • src/prepare_and_split_data.ipynb: Experimenteller Code um Sensordaten bereinigen
    • src/calculate_slope_per_station.py/.ipynb: Füllgeschwindigkeiten von bereinigten Sensordaten berechnen.
    • src/calc_prediction_accuracy.py: Berechnen des Standard Errors mit Testdaten.
  • Zusammenhang von Füllgeschwindigkeit und Bevölkerungsdichte
    • src/get_population_per_collection_point.py: Zuweisen von Quartier/Bevölkerung und Sammelstelle
    • src/calculate_slope_with_population.py: Untersuchung von Zusammenhang

Resultate

Diagramme und Grafiken in img/ Ordner. Füllgeschwindigkeiten und weitere statistische Werte in results/*_slopes.csv.

Infos

Team

  • Dominik Gschwind, @N3xed
  • Cedric Christen, @cedi4155476
  • Roman Weiss, @romanweiss123
  • Roméo Bornand
This content is a preview from an external site.
 

Joined the team

1 year ago ~ romeo_bornand

Project

Event finish

Repository updated

1 year ago ~ gaston_wey

Edited (version 26)

1 year ago ~ gaston_wey

Edited (version 24)

1 year ago ~ n3xed

Edited (version 23)

1 year ago ~ n3xed

Edited (version 22)

1 year ago ~ n3xed

Edited (version 21)

1 year ago ~ cdric_christen

Edited (version 20)

1 year ago ~ cdric_christen

Edited (version 19)

1 year ago ~ cdric_christen

Remove image from readme (@N3xed)

Edited (version 18)

1 year ago ~ cdric_christen

Project

Update readme (@N3xed)

Edited (version 17)

1 year ago ~ cdric_christen

Project

Edited (version 16)

1 year ago ~ cdric_christen

Edited (version 15)

1 year ago ~ cdric_christen

Update README, add results (@N3xed)

Edited (version 14)

1 year ago ~ cdric_christen

Project

Edited (version 12)

1 year ago ~ cdric_christen

Edited (version 10)

1 year ago ~ cdric_christen

Edited (version 9)

1 year ago ~ cdric_christen

Edited (version 8)

1 year ago ~ cdric_christen

Edited (version 7)

1 year ago ~ roman_weiss

Edited (version 6)

1 year ago ~ cdric_christen

Edited (version 5)

1 year ago ~ cdric_christen

Edited (version 3)

1 year ago ~ roman_weiss

Edited (version 2)

1 year ago ~ cdric_christen

Joined the team

1 year ago ~ n3xed

Repository updated

1 year ago ~ cdric_christen

Joined the team

1 year ago ~ cdric_christen

Challenge shared
Tap here to review.

1 year ago ~ cdric_christen

Create slides (@N3xed)

save shape files of geometries (@cedric.christen)

Add plots (@N3xed)

Merge branch 'master' of github.com:N3xed/glass-utilization-predictor merge (@cedric.christen)

population correlation plot (@cedric.christen)

Remove unused python file (@N3xed)

Calculate slope in one model, add graphs (@N3xed)

prototype ready without graphics (@cedric.christen)

Fix start time of accuracy calc (@N3xed)

calculate slope pop prototype (@cedric.christen)

Add accuracy calc script (@N3xed)

Add script for preparing the sensor data and splitting it into periods (@N3xed)

get population per collection point (@cedric.christen)

Füllstandanalyse (@cedric.christen)

Add people and districts jupyter notebook (@N3xed)

Add reame, datasets.md (@N3xed)

Start

 
Contributed 1 year ago by cdric_christen for Open Data Hack St.Gallen
All attendees, sponsors, partners, volunteers and staff at our hackathon are required to agree with the Hack Code of Conduct. Organisers will enforce this code throughout the event. We expect cooperation from all participants to ensure a safe environment for everybody.

Creative Commons LicenceThe contents of this website, unless otherwise stated, are licensed under a Creative Commons Attribution 4.0 International License.