Guest User

Untitled

a guest
Jul 21st, 2018
82
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 0.64 KB | None | 0 0
  1. # Using the official tensorflow serving image from docker hub as base image
  2. FROM tensorflow/serving
  3.  
  4. # Installing NGINX, used to rever proxy the predictions from SageMaker to TF Serving
  5. RUN apt-get update && apt-get install -y --no-install-recommends nginx git
  6.  
  7. # Copy our model folder to the container
  8. COPY saved_model_half_plus_three /saved_model_half_plus_three
  9.  
  10. # Copy NGINX configuration to the container
  11. COPY nginx.conf /etc/nginx/nginx.conf
  12.  
  13. # starts NGINX and TF serving pointing to our model
  14. ENTRYPOINT service nginx start | tensorflow_model_server --rest_api_port=8501 \
  15. --model_name=half_plus_three \
  16. --model_base_path=/saved_model_half_plus_three
Add Comment
Please, Sign In to add comment