Recipe Recommender System

This is a recipe recommender built off of Jamie Oliver's recipes on https://www.jamieoliver.com/ This application allows the user to enter n amount of ingredients, their macronutritional and caloric minimums, which will then output a dataframe that meets their requirements.

Libraries required (Python)

• NumPy
• Pandas
• Matplotlib
• Seaborn
• Scikit-learn
• re
• nltk
• wordcloud
• datetime
• unidecode
• string
• ast
• bs4
• time
• requests
• streamlit

Streamlit

Run the .py file in streamlit using VSCode terminal with the libraries above installed. To run the streamlit app, use:

streamlit run modelling2.py

Jupyter Notebook

For jupyter notebook followthrough, run the following notebooks in this order:

1. WebscrapingJO.ipynb (Keep in mind, this process will take 40-50 minutes depending on your system, so i suggest installing the Jamie_Oliver_Recipes.csv and move onto step 2, or install Cleaned_data.csv and skip to step 3).

2. NLP_DC.ipynb

3. Modelling.ipynb

Inspirations

This project was inspired by:

• https://towardsdatascience.com/building-a-recipe-recommendation-system-297c229dda7b

License

GNU General Public License version 3