From 7097ddac6bcec6ab79e68c34951757c41f8100e0 Mon Sep 17 00:00:00 2001 From: MaoXianxin Date: Sun, 25 Jul 2021 22:41:35 +0800 Subject: [PATCH] tensorflow serving with docker --- tensorflow serving with docker.md | 33 +++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) create mode 100644 tensorflow serving with docker.md diff --git a/tensorflow serving with docker.md b/tensorflow serving with docker.md new file mode 100644 index 0000000..0a75a68 --- /dev/null +++ b/tensorflow serving with docker.md @@ -0,0 +1,33 @@ +本次教程的目的是带领大家看下如何用 Docker 部署深度学习模型的 + +第一步我们需要 pull 一个 docker image + +```bsh +sudo docker pull tensorflow/serving +``` + +![](https://maoxianxin1996.oss-accelerate.aliyuncs.com/codechina1/20210725223155.png) + +如上图所示，执行 pull 之后，我们看到本地已经存在 tensorflow/serving:latest + +接下来我们 clone 一个仓库 + +```bsh +git clone https://github.com/tensorflow/serving +``` + +![](https://maoxianxin1996.oss-accelerate.aliyuncs.com/codechina1/20210725223402.png) + +上图中的 saved_model_half_plus_two_cpu 就是我们想要部署的模型 + +然后我们可以直接运行以下命令实现部署 + +![](https://maoxianxin1996.oss-accelerate.aliyuncs.com/codechina1/20210725223549.png) + +运行结果如下图所示，我们可以看到 Exporting HTTP/REST API at:localhost:8501，那么就代表着部署成功了 + +![](https://maoxianxin1996.oss-accelerate.aliyuncs.com/codechina1/20210725223630.png) + +接下来我们可以进行预测，返回的结果也能对的上 + +![](https://maoxianxin1996.oss-accelerate.aliyuncs.com/codechina1/20210725223742.png) \ No newline at end of file -- GitLab