Accelerating TensorFlow 1.13.0-rc0 With TensorRT 5.0.2 Using The 19.02 Container

These release notes are for accelerating TensorFlow 1.13.0-rc0 with TensorRT version 5.0.2 using the TensorFlow 19.02 container. For specific details about TensorRT, see the TensorRT 5.0.2 Release Notes.

Key Features And Enhancements

This release includes the following key features and enhancements.
TensorFlow operators
The following operators can now be converted from TensorFlow to TensorRT: ExpandDims, Reshape, Sigmoid, Sqrt, Square, Squeeze, StridedSlice and Tanh. For more information, see Supported Ops.
INT8 Pre-quantization
You can manually insert quantization ranges (generated during quantization-aware training) to the graph, and then TF-TRT can use them during INT8 inference. That means calibration is not required with this feature. For more information, see INT8 Quantization.

Compatibility

Limitations Of Accelerating TensorFlow With TensorRT

  • TF-TRT is not supported in the TensorRT containers.

Deprecated Features

  • Support for TensorRT 3 has been removed.