Faster Transformer 3.0 编码器的 INT8 量化实现.pdf
50墨值下载
Yu Chen, 2020/12
THE INT8 QUANTIZATION
OF FASTER TRANSFORMER
3.0 ENCODER
2
What is Faster Transformer
Introduce the Faster Transformer & its Encoder
What is INT8 Quantization
Introduce the INT8 quantization technique used in Faster Transformer
3.0 Encoder
How to do INT8 Quantization with
cuBLASLt
Introduce how to use
cuBLASLt to implement INT8 Quantization
INT8 Quantization of Faster Transformer Encoder
The performance of Faster Transformer INT8 Encoder
Demonstrate the performance
Further improvement
INT8 output GEMM
Summary
AGENDA
of 36
50墨值下载
【版权声明】本文为墨天轮用户原创内容，转载时必须标注文档的来源（墨天轮），文档链接，文档作者等基本信息，否则作者和墨天轮有权追究责任。如果您发现墨天轮中有涉嫌抄袭或者侵权的内容，欢迎发送邮件至：contact@modb.pro进行举报，并提供相关证据，一经查实，墨天轮将立刻删除相关内容。
下载排行榜
文档被以下合辑收录
评论