Pinned[ML] bitsandbytes NF4 quantize, dequantize analysisAnalyzing bitsandbytes quantize & dequantize process with real examplesSep 16, 2023Sep 16, 2023
[NLG] Outlines — Efficient Guided Generation for Large Language Models (Willard & Louf 2023)Analysis of paper “Efficient Guided Generation for Large Language Models (Willard & Louf 2023)”[1] which describes the methodology of…Aug 20Aug 20
[IR] (SIGIR 21) SPLADE — Sparse Lexical and Expansion Model for First Stage RankingNAVER Labs paper https://arxiv.org/abs/2107.05720May 13May 13
Apple Silicon Experiment 6— MLX Framework BERT Port [WIP]Apple announced MLX Framework for ML ModelsDec 7, 2023Dec 7, 2023
Apple Silicon Experiment 5— WWDC23 torch 2.1 mps updatesWith WWDC 2023 and macOS Sonoma beta, there were a few key updates to using PyTorch on the mac.Jun 10, 20231Jun 10, 20231
Apple Silicon Experiment 4 — PyTorch mps backendhttps://pytorch.org/blog/introducing-accelerated-pytorch-training-on-mac/May 22, 2022May 22, 2022
Apple Silicon Experiment 3 — tensorflow-metalMetal accelerated Tensorflow is now available as a part of macOS 12 beta, so I decided to try it out.Oct 16, 2021Oct 16, 2021
Apple Silicon Experiment 2 — TensorflowI’ve tried 2 methods of using tensorflow python package on Apple SiliconFeb 26, 2021Feb 26, 2021