https://github.com/andriikot/smartmatch

Smart Match with python
https://github.com/andriikot/smartmatch

Last synced: about 1 month ago
JSON representation

Smart Match with python

• Host: GitHub
• URL: https://github.com/andriikot/smartmatch
• Owner: AndriiKot
• Created: 2024-11-26T10:48:39.000Z (about 1 month ago)
• Default Branch: main
• Last Pushed: 2024-11-26T11:36:01.000Z (about 1 month ago)
• Last Synced: 2024-11-26T11:45:55.843Z (about 1 month ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 0 Bytes
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# SmartMatch

This project is designed to find the most similar

song based on the text provided by the user.

It utilizes a transformer model from the `transformers`

library for vectorizing song lyrics and computes

similarity using cosine similarity.

## Technologies:

  

    

      Python

      Jupyter

      Hugging Face

      PyTorch

      NumPy

      scikit-learn

    

  

  

    

      

        

          

        

      

      

        

          

        

      

      

        

          

        

      

      

        

          

        

      

      

        

          

        

      

      

        

          

        

      

    

  

## Contents

- [Functionality](#functionality)

- [Requirements](#requirements)

- [Installation](#installation)

- [Usage](#usage)

- [Notes](#notes)

## Functionality

The program allows you to:

- Load song lyrics from text files.

- Vectorize song lyrics using the `distilbert-base-uncased` model.

- Find the most similar song based on the user-provided text.

## Requirements

To run the program, you need to install the following libraries:

- `transformers`

- `torch`

- `scikit-learn`

## Installation

To install the necessary libraries, run the following command in your Jupyter Notebook:

```python

!pip install transformers torch scikit-learn

```

## Usage

1. **Data Preparation:**

   - Create a folder named `data` in the directory where your Jupyter Notebook is located.

   - Place text files containing the song lyrics in this folder. Make sure all files have a `.txt` extension.

2. **Running the Program:**

   - Open your Jupyter Notebook.

   - Copy and paste the provided program code into the cells of your notebook.

   - Execute the cells in the following order:

     - A cell for importing libraries and loading the song lyrics.

     - A cell for inputting the user text and searching for a similar song.

3. **Text Input:**

   - After running the code, enter the text for which you want to find a similar song when prompted.

   Example code for text input:

   ```python

   user_input = input("Please enter the text you want to find a similar song for: ")

   similar_song_name, similarity_score = find_similar_song(user_input)

   print(f"The most similar song is: {similar_song_name} with a similarity score of {similarity_score:.4f}")

   ```

## Notes

- If you do not have a Graphics Processing Unit (GPU), make sure that `PyTorch` is installed with CPU support, or simply use the CPU as your device.

- Depending on the volume of songs you have loaded and the power of your computer, vector calculation may take some time.

## License

This project is licensed under the MIT License. Please refer to the LICENSE file for more details.