Python codes and Jupyter notebooks to replicate the 99 pipelines and their empirical evaluations. The pipelines consist of combinations of fine-tuned and non-fine-tuned Feature Extraction methods, Clustering algorithms, and Dimensionality Reduction approaches.