sqlalchemy-flask-api-weather

Author: Erin James Wills, ejw.data@gmail.com

Photo by NASA on Unsplash

Overview

---

The purpose of this repo is to become familiarized with SQLAlchemy's Object Relational Mapper (ORM). To do this I will be using a SQLite database that has two tables about Hawaii weather (measurements and stations). The code associated with this project can be found in the analysis folder within a jupyter notebook named climate_analysis.ipynb. In addtion to analyzing the data, a Flask API was created that connects to the database and accepts five different API requests - three of the requests return a predefined query result but the other two API's allow for the user to include one or two user inputs that are used within the API request as database filters.

Technologies

• Python
• Pandas
• SQLAlchemy
• Flask
• SQLite

Data Source

Dataset generated by Trilogy Education Services. Origins beyond this is unknown.

Setup and Installation

1. Environment needs the following:
   • Python 3.6+
   • pandas
   • flask
   • sqlalchemy.ext.automap
   • sqlalchemy.orm
   • datetime
   • numpy
2. Activate your environment
3. Clone the repo to your local machine
4. Start Jupyter Notebook within the environment from the repo
5. Run climate_analysis.ipynb to test connections and queries
6. To run the flask app:
   • In terminal, navigate to the top-level repo folder
   • Activate your environment
   • In terminal run: python app.py
   • The routes that can be used are listed on the page.

Examples

Fig.1 - API route options displayed in a browser

Fig.2 - API results returning a single day's precipation and temperature measurements

Fig.3 - Matplotlib histogram of temperatures returned from a date range query

Fig.4 - Matplotlib stacked area chart of temperatures returned from a date range query