This is a gift recommendation system that, based on minimal input data, can recommend a gift for a person. This is a learning project. As a result, it is supposed to create a simple web page or telegram bot, where you can pick up a gift.
.
├───config
│ └───params.yaml # configuration parameters
├───data
│ ├───raw
│ │ └───ozon_data.csv # csv file with all products
│ └───processed
│ ├───train.csv # dataset for training
│ └───valid.csv # dataset for validation
├───src
│ └───parse.py # scrapes products from ozon.ru
├───bot
│ ├───config_data
│ │ └───config.py # configuration parameters for bot
│ ├───handlers
│ │ ├───admin_handlers.py # handlers admin
│ │ └───user_handlers.py # handlers user
│ ├───keyboards
│ │ └───main_keyboards.py # keyboards
│ ├───.env #
│ └───bot.py #
├───.gitignore
├───requirements.txt # all required dependencies
├───Gift recommendation system.ipynb.ipynb # trains the model on the training dataset,
│ contains class with the tuned model,
│ example how the inference works
└───README.md
The device must have python and git. Commands are executed from the terminal.
- Clone the repo — git clone https://github.com/tutsilianna/Gift_Recommendation_System;
- Install all dependencies — pip install -r requirements.txt;
- ...