This github project hosts a rookie's experiment to resolve naively a complex text classification problem by comparing 2 approaches:
- Deep Learning (DL)technique which has regained tremendous attraction from the Machine Learning community to address NLP problem
- Traditional Technique which is basically based on non-DL principle
It's organized into below folders:
- data contains training/external source/staging data consumed or produced by the learning pipelines
- notebooks stores all Jupyter notebooks
- pretrained_models refers to models which are available publicly and are consumed "as is" in the experiment
- reports holds the reports summarizing the NLP experiment
- temp misc temporary artifacts