Guolin Ke

TUPE. Transformer with Untied Positional Encoding (TUPE). Code of paper "Rethinking Positional Encoding in Language Pre-training". Improve existing models like BERT.

SphereAR. Implementation of "Hyperspherical Latents Improve Continuous-Token Autoregressive Generation"

boosting_tree_benchmarks. Python

deep-learning-benchmarks. Python

fused_ops. Cuda

pytorch-docker. Dockerfile

sentencepiece. Unsupervised text tokenizer for Neural Network-based text generation.

machinelearning. ML.NET is an open source and cross-platform machine learning framework for .NET.

nccl. Optimized primitives for collective multi-GPU communication

BetterChatGPT. An amazing UI for OpenAI's ChatGPT (Website + Windows + MacOS + Linux)

DeepSpeed. DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.

transformers. 🤗Transformers: State-of-the-art Natural Language Processing for Pytorch and TensorFlow 2.0.

tokenizers. 💥Fast State-of-the-Art Tokenizers optimized for Research and Production

fairseq. Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

rdkit. The official sources for the RDKit library

apex. A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

guolinke.

pytorch. Tensors and Dynamic neural networks in Python with strong GPU acceleration

azhpc-images. Scripts for installing HPC libraries, tools, to be used in Azure HPC images

category_encoders. A library of sklearn compatible categorical variable encoders

openfold. Trainable PyTorch reproduction of AlphaFold 2

recommenders. Best Practices on Recommendation Systems

github_metrics.

arxiv-sanity-preserver. Web interface for browsing, search and filtering recent arxiv submissions

sparsity. R package for loading/saving sparse matrices (dgCMatrix) to SVMlight (LibFM, xgboost, LightGBM, H2O, etc.)

claude-code-test.