Skip to content

Latest commit

 

History

History
13 lines (9 loc) · 412 Bytes

File metadata and controls

13 lines (9 loc) · 412 Bytes

Keras Flask Deployment

This is an example of deploying a Keras model using Flask for image classification. It is just the stock ImageNet VGG16 model with some abstraction around the model and image pre-processing. A convenient shell script is provided to launch the Flask app. You can submit the test image from command line:

curl -X POST -F 'image=@images/dog.jpg' http://localhost:5000/predict