This crate is a Rust port of Google's BERT create pretraining data.
Model implementations for TrustformeRS
Ready-to-use NLP pipelines and language models
A repository of deep-learning peptide property prediction models for mass spectrometry data
Pre-trained models and model zoo for ToRSh deep learning framework
Tokenizers for TrustformeRS
Provides an implementation of today's most used tokenizers, with a focus on performances and versatility.
Provides an implementation of today's most used tokenizers, with a focus on performances and versatility.
Transformer-as-rules: Self-attention and FFN layers as einsum expressions
AI/Human task management system with file-based storage
Model loading for RLX — config parsing, safetensors weights, graph builders
Split code into semantic chunks using tree-sitter