ts-app 0.1.0rc1

A simple dashboard app to interactively fit ARIMA models.

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.

1. Fetch the necessary files:

   git clone https://github.com/Tim-Abwao/time-series-app.git
   cd time-series-app
   

2. Create the virtual environment:

   python3 -m venv venv
   source venv/bin/activate
   pip install -U pip
   pip install -r requirements.txt
   

3. Start the app:

   You can use the convenient run.sh script, or waitress:

   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.

1. Fetch the necessary files, just as above:

   git clone https://github.com/Tim-Abwao/time-series-app.git
   cd time-series-app
   

2. 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.