Year: 2021
Type: preprint
Abstract: This paper presents a methodology to separate the quantization process from the hardware-specific model compilation stage via a pre-quantized deep learning model description in standard ONNX format. Separating the quantization process from the model compilation stage enables independent development. The methodology is expressive to convey hardware-specific operations and to embed key quantization parameters into a ONNX model which enables hardware/software co-design. Detailed examples are given for both MLP and CNN based networks, which can be extended to other networks in a straightforward fashion.
Source: arXiv (Cornell University)
Authors: Hanebutte, Ulf, Baldwin, Andrew, Durakovic, Senad, Filipovich, Igor, Chien-Chun, Chou, Adamowicz, Damian, Chickles, Derek, Hawkes, David
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3202076421