A simple dashboard app to interactively fit ARIMA models.
Project description
Time Series App
A simple web app to learn a little about Time Series analysis and forecasting.
You can create a sample, or upload a file, and interactively fit a time series model on it. To give it a try, click here...
Installation
The easiest way to install the app is from PyPI using:
pip install ts-app
You can then start it using the command
ts_app
or even
python -m ts_app
Press CTRL
+ C
to stop it.
You can also start the app from an interactive session:
>>> import ts_app
>>> ts_app.run_app()
Manual set up
1. Using a virtual environment
You'll need Python 3.8 and above. Packages used include statsmodels, flask, dash, pandas and NumPy.
-
Fetch the necessary files:
git clone https://github.com/Tim-Abwao/time-series-app.git cd time-series-app
-
Create the virtual environment:
python3 -m venv venv source venv/bin/activate pip install -U pip pip install -r requirements.txt
-
Start the app:
You can use the convenient
run.sh
script, orwaitress
:waitress-serve --listen=127.0.0.1:8000 ts_app:server
then browse to localhost:8000 to interact with the web app.
Afterwards, use
CTRL
+C
to stop it.
2. Using Docker
You'll need Docker.
-
Fetch the necessary files, just as above:
git clone https://github.com/Tim-Abwao/time-series-app.git cd time-series-app
-
Build an image for the app and run it in a container,
docker build --tag ts_app . docker run --name ts -d -p 8000:8000 --rm ts_app
in which case the app will be running at http://0.0.0.0:8000.
Afterwards, use
docker stop ts
to terminate it.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.