Provides LDA training pipeline, data preprocessing tools and visualization.
- modules/ (pipeline, vizualisation ...)
- img/ : images used in documentation en report
- 5 notebooks :
- genius-lyrics-topics-modeling.ipynb is the original notebook aim to test the several models.
- genius-song-lyrics-lda-1960-70.ipynb training pipeline from 1960 to 1979
- genius-song-lyrics-lda-1980-90.ipynb training pipeline from 1980 to 1999
- genius-song-lyrics-lda-2000-10.ipynb training pipeline from 2000 to 2010
- data_vizualisation.ipynb contain all usefull data visualisation implemented in modules/visualization.py