Phone Price Prediction

Problem Description

Being a Mobile Developer, I'm passionate about all things mobile. Hence, in this Midterm Project, the task to tackle is how to predict the price of a phone based on its features like: the Operating System (Android / iOS), the prestige of its brand (Apple, Samsung, Xiaomi, OPPO) and other technical specifications (battery, screen size, etc).

Data

The data used in this project is the Mobile Phone Specifications and Prices dataset, which can be found in Kaggle.

Here is a detailed look of the features provided in the dataset, with their descriptions and unit of measure (if applicable).

Note: Some feature names are different from those of the actual dataset, which are cleaned.

Getting Started

This is a set of instructions on setting up this project locally. To get a local copy up and running follow these simple example steps.

Prerequisites This is an example of how to list things you need to use this software

• Python
• Pipenv
• Docker
• Windows Subsystem for Linux (if using Windows)

Installing Dependencies

You can install the dependencies with pipenv, as they are specified in the Pipfile and Pipfile.lock, by running the following commands:

pipenv install
pipenv shell

Building the Model

You can run the train.py file or the full model_training.ipynb Jupyter Notebook to perform all the steps required to train the final model used in this project, which is a Gradient Boosted Trees Regressor (XGBoost).

So, in order to train the model, you can run the following command:

python train.py

Serving the Model (Locally)

We can serve our model with BentoML and the predict.py script by running:

bentoml serve predict.py:svc

This scripts loads the latest model available locally, which can be used in the browser as BentoML automatically creates a Swagger UI at http://localhost:3000. The variables expected by the model to predict the price of a phone can be found in the sample_record.json file.

Building the Bento and Containerizing the Model

To containerize the model into a Docker Container with all the required dependencies we use BentoML, which facilitates this procedure. With BentoML, in order to containerize the model, it's only necessary to specify a bentofile.yaml file which specifies the project name, owner and all the required dependencies. After that, we just have to run the command.

bentoml build

Similar to saving a model, a unique version tag will be automatically generated for the newly created Bento. The output expected is similar to the one shown below:

bentoml build

Building BentoML service "phone_price_predictor:dpijemevl6nlhlg6" from build context "/home/user/gallery/quickstart"
Packing model "iris_clf:zy3dfgxzqkjrlgxi"
Locking PyPI package versions..

██████╗░███████╗███╗░░██╗████████╗░█████╗░███╗░░░███╗██╗░░░░░
██╔══██╗██╔════╝████╗░██║╚══██╔══╝██╔══██╗████╗░████║██║░░░░░
██████╦╝█████╗░░██╔██╗██║░░░██║░░░██║░░██║██╔████╔██║██║░░░░░
██╔══██╗██╔══╝░░██║╚████║░░░██║░░░██║░░██║██║╚██╔╝██║██║░░░░░
██████╦╝███████╗██║░╚███║░░░██║░░░╚█████╔╝██║░╚═╝░██║███████╗
╚═════╝░╚══════╝╚═╝░░╚══╝░░░╚═╝░░░░╚════╝░╚═╝░░░░░╚═╝╚══════╝

Successfully built Bento(tag="phone_price_predictor:dpijemevl6nlhlg6")

After creating the Bento, we can finally containerize the model by running

bentoml containerize phone_price_predictor:latest

which will create a Docker image that we can check by running docker image ls.

We can run this image by passing phone_price_predictor:f35knqlbck3zlhfw to "docker run". For example: "docker run -it --rm p 3000:3000 phone_price_predictor:f35knqlbck3zlhfw"

Note: Naturally, the tag can vary