[MXNET-503] Website landing page for MMS #11037
Conversation
docs/mms/index.md
Outdated
| @@ -0,0 +1,91 @@ | |||
| # Model Server for Apache MXNet (incubating) | |||
|
|
|||
| Model Server for Apache MXNet (incubating), otherwise known as MXNet Model Server (MMS), is an open source project aimed at providing a simple yet scalable solution for model inference. It is a set of command line tools for packaging model archives and serving them. The tools are written in Python, and have been extended to support containers for easy deployment and scaling. MMS also supports basic logging and advanced logging with AWS CloudWatch integration. | |||
There was a problem hiding this comment.
Should we link the Github repo from MMS?
docs/mms/index.md
Outdated
| @@ -0,0 +1,91 @@ | |||
| # Model Server for Apache MXNet (incubating) | |||
|
|
|||
| Model Server for Apache MXNet (incubating), otherwise known as MXNet Model Server (MMS), is an open source project aimed at providing a simple yet scalable solution for model inference. It is a set of command line tools for packaging model archives and serving them. The tools are written in Python, and have been extended to support containers for easy deployment and scaling. MMS also supports basic logging and advanced logging with AWS CloudWatch integration. | |||
There was a problem hiding this comment.
Word "metrics" is important along with logging.
docs/mms/index.md
Outdated
| @@ -0,0 +1,91 @@ | |||
| # Model Server for Apache MXNet (incubating) | |||
|
|
|||
| Model Server for Apache MXNet (incubating), otherwise known as MXNet Model Server (MMS), is an open source project aimed at providing a simple yet scalable solution for model inference. It is a set of command line tools for packaging model archives and serving them. The tools are written in Python, and have been extended to support containers for easy deployment and scaling. MMS also supports basic logging and advanced logging with AWS CloudWatch integration. | |||
There was a problem hiding this comment.
Should we say MMS supports both Gluon and MXNet symbolic models along with ONNX models?
docs/mms/index.md
Outdated
|
|
||
| While the name implies that MMS is just for MXNet, it is in fact much more flexible, as it can support models in the [ONNX](https://onnx.ai) format. This means that models created and trained in PyTorch, Caffe2, or other ONNX-supporting frameworks can be served with MMS. | ||
|
|
||
| To find out more about MXNet's support for ONNX models and using ONNX with MMS, refer to the following resources: |
There was a problem hiding this comment.
There are some tutorials on exporting PyTorch models to ONNX, should we link it from the work PyTorch models?
| mxnet-model-server \ | ||
| --models squeezenet=https://s3.amazonaws.com/model-server/models/squeezenet_v1.1/squeezenet_v1.1.model | ||
| ``` | ||
|
|
There was a problem hiding this comment.
Should we show an example output here for clarity of URL, port number etc. that we use in next section? Something like - Your server should start up like below etc.
There was a problem hiding this comment.
added output sample
| * [MXNet Forum - MMS Discussions](https://discuss.mxnet.io/c/mxnet-model-server) | ||
|
|
||
|
|
||
| ## Further Reading |
There was a problem hiding this comment.
Probably we should also reference AWS blogpost on using MMS with Fargate for productionization of DL model?
There was a problem hiding this comment.
added link to blog
|
Pinging @piiswrong for review |
| * [Create an ONNX model (that will run with MMS) from PyTorch](https://github.com/onnx/onnx-mxnet/blob/master/README.md#quick-start) | ||
|
|
||
|
|
||
| ## Using Containers |
There was a problem hiding this comment.
Should we point users to the Docker Hub builds?
https://hub.docker.com/r/awsdeeplearningteam/mms_cpu and https://hub.docker.com/r/awsdeeplearningteam/mms_gpu
There was a problem hiding this comment.
I added a docker pull command and a link to the parent on docker hub. I think it would be nicer to have an actual organization page and have the repos listed there. One stop shopping, but this should be fine for now!
|
aaronmarkham
changed the title
|
Thanks @john-andrilla - I've addressed each comment in my latest commit (fixed stuff, removed "free"). |
This reverts commit d552640.
* new landing page for MMS * added links * more links * updates from feedback * added docker hub info * addressed comments
…che#11154) This reverts commit d552640.
* new landing page for MMS * added links * more links * updates from feedback * added docker hub info * addressed comments
…che#11154) This reverts commit d552640.
Description
This is the content for a website landing page for MMS. Navigation updates will come later.