https://github.com/thomasjungblut/tfs-inception

Tensorflow Serving with an inception model
https://github.com/thomasjungblut/tfs-inception

Last synced: 3 months ago
JSON representation

Tensorflow Serving with an inception model

• Host: GitHub
• URL: https://github.com/thomasjungblut/tfs-inception
• Owner: thomasjungblut
• License: apache-2.0
• Created: 2018-07-18T10:11:06.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2019-05-15T16:41:22.000Z (about 6 years ago)
• Last Synced: 2025-02-09T18:17:45.734Z (5 months ago)
• Homepage:
• Size: 6.84 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
[![Docker Repository on Quay](https://quay.io/repository/thomasjungblut/tfs-inception/status "Docker Repository on Quay")](https://quay.io/repository/thomasjungblut/tfs-inception)

# Tensorflow Serving

This docker image contains Tensorflow Serving (TFS) 1.8 with GPU support and a ready to run inceptionv3 model. 

The resulting image can be used in conjunction with the helm chart from [Tensorflow Inception Model Chart

](https://github.com/kubernetes/charts/tree/master/incubator/tensorflow-inception). 

With plain docker you can run via:

```bash

docker run -i --expose=9000 quay.io/thomasjungblut/tfs-inception tensorflow_model_server --port=9000 --model_name=inception --model_base_path=/serving/inception-export

```

For GPU support you can either use `nvidia-docker` or the `nvidia` runtime.