The notebook below follows our recommended inference workflow. This example demonstrates how to do model inference using TensorFlow with pre-trained ResNet-50 model and TFRecords as input data. In order to understand the following example, you need to understand how to do the following:
BinaryType is supported by Arrow-based conversion in PyArrow 0.10.0 and above. If you run on Databricks Runtime 5.3 ML (Unsupported) or above, you do not need to install PyArrow. Databricks Runtime 4.1-5.2 ML contains PyArrow 0.8.0. On these runtimes you must install PyArrow 0.10.0 or above as a PyPI Databricks library.