提交 424d0667 编写于 作者: W wangjiawei04

move docs

上级 e372964d
......@@ -82,4 +82,3 @@ read samples from data.txt, print results at the standard output.
We tested the performance of Bert-As-Service based on Padde Serving based on V100 and compared it with the Bert-As-Service based on Tensorflow. From the perspective of user configuration, we used the same batch size and concurrent number for stress testing. The overall throughput performance data obtained under 4 V100s is as follows.
![4v100_bert_as_service_benchmark](4v100_bert_as_service_benchmark.png)
......@@ -83,4 +83,3 @@ cat data.txt | python bert_client.py
我们基于V100对基于Padde Serving研发的Bert-As-Service的性能进行测试并与基于Tensorflow实现的Bert-As-Service进行对比,从用户配置的角度,采用相同的batch size和并发数进行压力测试,得到4块V100下的整体吞吐性能数据如下。
![4v100_bert_as_service_benchmark](4v100_bert_as_service_benchmark.png)
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册