Project document | SGI website | Justin's geometry course
books/
alice_wonderland/
raw.txt
processed.json
paragraph_scores.json
embeddings/
bge-m3/
qwen3-embedding/
e5-large-v2/
Run preprocess/preprocess.py to split the chapters and paragraphs in the book.
Run preprocess/get_embeddings.py to obtain the text embeddings for each paragraph.
Run preprocess/get_annotations.py to get the per-paragraph scores annotated by Qwen2.5-7B-Instruct.
Download the preprocessed files: link.
Implement geometric algorithms in geometry/.
Implement visualization scripts in visualization/.