Introduction
Last updated
Was this helpful?
Last updated
Was this helpful?
For a data scientist, the should be considered both the operations environment, once a model is released into production, and it also holds a long range of devloper friendly API's that can assist in the development of a new deep learning model.
Most notably the platform
Manages access to data
Provides a series of tools for visualization of results, predictions and images
Configures and tracks experiments and gives a managed artifact repository
Provides a seamless switch to operations/production once the model design phase is over
The documentation has three main sections detailing the platform REST interfaces, and python code that can be used for task automation. It details the training job API's and features, and, finally, there are a number of tutorials that may assist in getting familiar with the python package and a public repository with a small application sample .