1 个仓库
Splitting text by category (letters, numbers, punctuation) before tokenization to prevent cross-category merges.
Distinct from Recursive Text Splitting: Distinct from Recursive Text Splitting: splits by character category for BPE preprocessing, not recursive chunking for token limits.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Category-Based Splitters. Refine with filters or upvote what's useful.
Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.
Splits text by character category before tokenization to prevent cross-category BPE merges.