gauravchl / tensorflow_model_server

Dockerized tensorflow_model_server

tensorflow_model_server

Dockerized tensorflow_model_server

Pull Image

docker pull gauravchl/tensorflow_model_server

Run Container

docker run -d \
  --name tensorflow_model_server \
  -v ${PATH_TO_SAVED_MODEL}:/model \
  -e MODEL_NAME=${MODEL_NAME} \
  -p 9000:9000 \
  gauravchl/tensorflow_model_server

• Above command will start docker container named 'tensorflow_model_server' and will be accessed from port 9000

• Run following command to show the logs: docker logs -f tensorflow_model_server

• Make sure to replace ${PATH_TO_SAVED_MODEL} and ${MODEL_NAME} with your own data, Example:

docker run -d \
  --name tensorflow_model_server \
  -v /Users/om/imagenet-model/:/model \
  -e MODEL_NAME=imagenet \
  -p 9000:9000 \
  gauravchl/tensorflow_model_server

Stop Container

docker stop tensorflow_model_server

Remove Container

docker rm tensorflow_model_server

Build Image

# Build locally from tensorflow_model_server/
docker build -t gauravchl/tensorflow_model_server .

# Build from repo
docker build github.com/gauravchl/tensorflow_model_server