The notebook below is the second of six notebooks demonstrating how to perform distributed training with TensorFlowOnSpark on the MNIST dataset. Download the full set of notebooks or see the TensorFlowOnSpark guide for more information.
This notebook demonstrates how to construct a TensorFlow graph for distributed model training, and depends solely on TensorFlow. The helper functions defined in the notebook will be wrapped using TensorFlowOnSpark in a later step.
The next stage of the TensorFlowOnSpark training pipeline is to define the constants used in model training and evaluation. For more information, see Model Training and Evaluation Constants.