Deploying and scoring a deep learning model using frameworks
After a trained model has been identified for serving, you can deploy it by using the deployment and scoring system of IBM Watson Machine Learning. The deployed model can then be used to perform online scoring.
For guidelines on how to save a deep learning model in a format that is compatible with the deployment and scoring service please refer to the following documents:
- Deploying and scoring a deep learning TensorFlow model
- Deploying and scoring a deep learning Keras model
- Deploying and scoring a deep learning Caffe model
It is possible to perform many of these functions in different environments depending on your preference. For example, the Command Line Interface (CLI) or the Watson Machine Learning Python client can be used for deploying and scoring the models.