Hybridized On-Premise and Cloud (HOPC) Deployment Experimentation with Bfloat16
-
Updated
Apr 5, 2021
Hybridized On-Premise and Cloud (HOPC) Deployment Experimentation with Bfloat16
Comparison of PageRank algorithm using various datatypes.
Customizable floating point types, with all standard floating point operations implemented from scratch.
Comparison of vector element sum using various data types.
Basic linear algebra routines implemented using the chop rounding function
A Pytorch implementation of stochastic addition.
IEEE 754-style floating-point converter
A JAX implementation of stochastic addition.
A LLaMA2-7b chatbot with memory running on CPU, and optimized using smooth quantization, 4-bit quantization or Intel® Extension For PyTorch with bfloat16.
Floating-Point Arithmetic Library for Z80
Round matrix elements to lower precision in MATLAB
Half-precision floating point types f16 and bf16 for Rust.
oneAPI Deep Neural Network Library (oneDNN)
Add a description, image, and links to the bfloat16 topic page so that developers can more easily learn about it.
To associate your repository with the bfloat16 topic, visit your repo's landing page and select "manage topics."