Number-of-Vehicles-Prediction-in-Besiktas-Using-Neural-Networks

App Link : https://numberofvehiclesprediction.herokuapp.com

Traffic Data : https://data.ibb.gov.tr/dataset/saatlik-trafik-yogunluk-veri-seti (2020.01.01 - 2020.12.27)

Weather Data : https://www.worldweatheronline.com

It is a little project that aims to predict the number of vehicles in different places of Besiktas,Istanbul.

Coordinates Used

streamlit
tensorflow-cpu
keras
folium
pandas
numpy
sklearn
matplotlib
seaborn
plotly
streamlit-folium
wwo-hist

Slug size was more than 500 MB so I used tensorflow-cpu to decrease it. Maybe I can deploy the project with Docker later.

runtime.txt

python-3.7.10

License MIT

Name		Name	Last commit message	Last commit date
Latest commit History 62 Commits
.gitignore		.gitignore
.slugignore		.slugignore
Besiktas.csv		Besiktas.csv
LICENSE		LICENSE
Procfile		Procfile
README.md		README.md
app.py		app.py
capture.gif		capture.gif
get_data.ipynb		get_data.ipynb
region_1.csv		region_1.csv
region_10.csv		region_10.csv
region_11.csv		region_11.csv
region_12.csv		region_12.csv
region_13.csv		region_13.csv
region_14.csv		region_14.csv
region_15.csv		region_15.csv
region_16.csv		region_16.csv
region_2.csv		region_2.csv
region_3.csv		region_3.csv
region_4.csv		region_4.csv
region_5.csv		region_5.csv
region_6.csv		region_6.csv
region_7.csv		region_7.csv
region_8.csv		region_8.csv
region_9.csv		region_9.csv
requirements.txt		requirements.txt
runtime.txt		runtime.txt
setup.sh		setup.sh