From be8aa2c02465298fe94f4754619bda79abb3c3a7 Mon Sep 17 00:00:00 2001 From: Beth Dean Date: Thu, 6 Jun 2019 07:04:00 -0700 Subject: [PATCH] Add link to quantization tools for model conversion (#529) --- source/clear-linux/tutorials/dlrs/dlrs.rst | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/source/clear-linux/tutorials/dlrs/dlrs.rst b/source/clear-linux/tutorials/dlrs/dlrs.rst index 7af186c9..7d51ba4b 100644 --- a/source/clear-linux/tutorials/dlrs/dlrs.rst +++ b/source/clear-linux/tutorials/dlrs/dlrs.rst @@ -137,6 +137,9 @@ TensorFlow. You can replace the model with one of your choice supported by the TensorFlow benchmarks. + If you are using an FP32 based model, it can be converted to an int8 model using `Intel® quantization tools`_ + + PyTorch single and multi-node benchmarks **************************************** @@ -444,3 +447,5 @@ Related topics .. _Jupyter Notebook: https://jupyter.org/ .. _Release notes on Github\*: https://github.com/clearlinux/dockerfiles/blob/master/stacks/dlrs/releasenote.md + +.. _Intel® quantization tools: https://github.com/IntelAI/tools/blob/master/tensorflow_quantization/README.md#quantization-tools