paper-with-me

홈 › Papers

QPyTorch: A Low-Precision Arithmetic Simulation Framework

2019-10-09 · Tianyi Zhang, Zhiqiu Lin, Guandao Yang, Christopher De Sa

Low-precision training reduces computational cost and produces efficient models. Recent research in developing new low-precision training algorithms often relies on simulation to empirically evaluate the statistical effects of quantization while avoiding the substantial overhead of building specific hardware. To support this empirical research, we introduce QPyTorch, a low-precision arithmetic simulation framework. Built natively in PyTorch, QPyTorch provides a convenient interface that minimizes the efforts needed to reliably convert existing codes to study low-precision training. QPyTorch is general, and supports a variety of combinations of precisions, number formats, and rounding options. Additionally, it leverages an efficient fused-kernel approach to reduce simulator overhead, which enables simulation of large-scale, realistic problems. QPyTorch is publicly available at https://github.com/Tiiiger/QPyTorch.

📄 PDF Abstract BibTeX arXiv:1910.04540

Code (2)

Tiiiger/QPyTorch 공식 구현 pytorch
robust-systems-group/illusion_system

Tasks

Quantization

Similar Papers 제목 키워드 기반

Neural Quantum States in Mixed Precision

2026-01-28 · Massimo Solinas, Agnes Valenti, Nawaf Bou-Rabee, Roeland Wiersema arxiv

Scientific computing has long relied on double precision (64-bit floating point) arithmetic to guarantee accuracy in simulations of real-world phenomena. However, the growing availability of hardware accelerators such as…

Deep Neural Networks to Correct Sub-Precision Errors in CFD

2022-02-09 · Akash Haridas, Nagabhushana Rao Vadlamani, Yuki Minamoto

Information loss in numerical physics simulations can arise from various sources when solving discretized partial differential equations. In particular, errors related to numerical precision ("sub-precision errors") can …

Revisiting MUSIC: A Finite-Precision Perspective

2025-03-16 · Yiming Fang, Li Chen, Ang Chen, Weidong Wang

The high computational complexity of the multiple signal classification (MUSIC) algorithm is mainly caused by the subspace decomposition and spectrum search, especially for frequent real-time applications or massive sens…

INT-FP-QSim: Mixed Precision and Formats For Large Language Models and Vision Transformers

2023-07-07 · Lakshmi Nair, Mikhail Bernadskiy, Arulselvan Madhavan, Craig Chan 외

The recent rise of large language models (LLMs) has resulted in increased efforts towards running LLMs at reduced precision. Running LLMs at lower precision supports resource constraints and furthers their democratizatio…

Quantization

Bitwidth-Specific Logarithmic Arithmetic for Future Hardware-Accelerated Training

2025-10-20 · Hassan Hamad, Yuou Qiu, Peter A. Beerel, Keith M. Chugg arxiv

While advancements in quantization have significantly reduced the computational costs of inference in deep learning, training still predominantly relies on complex floating-point arithmetic. Low-precision fixed-point tra…