Skip to content

Flask App Project

Code Structure


Top level

  • app.py: Applications entrypoint. Used to start the webserver. Here we register our blueprints
  • airbnb_base_logger.py: Logger for incoming information


Controllers

One file per logical unit:

  • ping: Typical ping endpoint.
  • stats: Declares the endpoint where we can request the statistics.
  • models: Accepts data for an individual Airbnb listing and returns a predicted price.
  • airbnb: Supports basic HTTP requests. GET, POST (to be implemented), DELETE an airbnb based on id.


Repo

The data storage of the application. Should include the airbnb.csv initially. In case you delete or update something on the dataset, you can always replace it with the original_airbnb_copy.csv and bring it to its original state. Here you can find and the trained-model that runs in the background for the price prediction


Services

The business logic of our app.

  • stats: Functions that calculate several stats based on aggregations by utilized the pandas lib.
  • models: Functions that perform all the preprocessing on the input data and use the pre-trained model to predict the price of the listing.
  • aibnb: Mappings, utilities and anything else needed to work with the airbnb resource.


Test

Execute basic tests for the app.

python -m pytest

Execute stress tests 

pip install locust uvicorn
python app.py
locust --host http://localhost:5000   -f tests/stress/benchmark.py

or run test resuls on the terminal. Ex. 10 users with spanning rate 5.

locust --host http://localhost:5000   -f tests/stress/benchmark.py --headless --host http://localhost:5000 -u 10 -r 5

From here you get a user interface. It's typically available at 0.0.0.0:8089 unless you've changed the port number.