• My Content : Books recommandation mobile app
  • Goals
  • Installation
    • Prerequisites
    • Virtual environment
    • Dependencies
    • Environment variables
    • Azure resources
  • Usage
    • Download data
    • Run Notebook
    • Quality Assurance
  • Troubleshooting

---

My Content : Books recommandation mobile app

Repository of OpenClassrooms' AI Engineer path, project #9

Goal : use Azure Machine Learning and Azure Functions services, a Recommander system embedded in a React-Native mobile app to produce the MVP of a books recommandation mobile app.

You can see the results here :

• Presentation
• Notebook : HTML page with interactive plots

This repository is part of a 3-repos project :

• Main repo : My Content : Books recommandation mobile app : this repo
• Azure Function
• Mobile App

Goals

• create a first Recommder System based on Content-Based Filtering
  • Main notebook (EDA)
  • Content-Based Filtering
• improve the Recommder System with Collaborative Filtering
  • Collaborative Filtering with Surprise
  • Collaborative Filtering with Implicit
• improve the Recommder System with Hybrid Filtering
  • Hybrid Filtering with LightMF
• create the React-Native mobile app
  • Mobile App
• integrate the Recommder System in Azure Functions
  • Store recommendations in Azure CosmosDB
  • Azure Function

Installation

Prerequisites

• Python 3.9

Virtual environment

# python -m venv env
# > or just :
make venv
source env/bin/activate

Dependencies

# pip install kaggle jupyterlab ipykernel ipywidgets widgetsnbextension graphviz python-dotenv requests matplotlib seaborn plotly numpy statsmodels pandas sklearn transformers tensorflow
# > or :
# pip install -r requirements.txt
# > or just :
make install

Environment variables

• Set environment variable values in .env file (copy or rename .env.example).

Azure resources

The app will query an Azure CosmosDB database and an Azure Function that will return a list of relevant articles.

• Azure Cosmos DB input binding for Azure Functions 2.x and higher - HTTP trigger, look up ID from route data
• Quickstart: Create a function in Azure with Python using Visual Studio Code

Usage

Download data

Download, extract and upload to Azure Cityscape zip files.

make dataset

Run Notebook

jupyter-lab notebooks/main.ipynb

Quality Assurance

# make isort
# make format
# make lint
# make bandit
# make mypy
# make test
# > or just :
make qa

Troubleshooting

• Fix Plotly issues with JupyterLab

cf. Plotly troubleshooting

jupyter labextension install jupyterlab-plotly

• If using Jupyter Notebook instead of JupyterLab, uncomment the following lines in the notebook

import plotly.io as pio
pio.renderers.default='notebook'