Model evaluation code is extracted here in order to create a separate inference mode for the project. The evaluation code is then embedded into a flask app that accepts API requests. There are two docker files:
- dockerfile.cpu - creates a CPU bonud container
- dockerfile.gpu - creates a GPU bound container
The app uses Azure Blob Storage as an image repository as well as Application Insights for logging telemetry.
There is a three step process running the application and generating bird images.
- Create the container (optionally choose the cpu or gpu dockerfile:
docker build -t "attngan" -f dockerfile.cpu . - Run the container (replace the key's with the appropriate blob storage location as well as App Insights Key):
docker run -it -e BLOB_KEY=KEY -e TELEMETRY=TELEMETRY_KEY -p 5678:8080 attngan - Call the API:
curl -H "Content-Type: application/json" -X POST -d '{"caption":"the bird has a yellow crown and a black eyering that is round"}' http://locahost:5678/api/v1.0/bird
You should have your very own image generator.