From 45ce94d4df90ebffcdd1c03a0a76d7e660327764 Mon Sep 17 00:00:00 2001 From: Beth Dean Date: Fri, 28 Jun 2019 10:46:25 -0700 Subject: [PATCH] Add link for launcher.py and note MKL HW reqs for DLRS tutorial. (#570) * Add link for launcher.py and note MKL HW reqs for DLRS tutorial. * Reword note about MKL, change from note to warning. Closes #560, Closes #559 * Add note specifying Kubeflow v0.4 when using launcher.py * Closes #559 , Closes #560 --- source/clear-linux/tutorials/dlrs/dlrs.rst | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/source/clear-linux/tutorials/dlrs/dlrs.rst b/source/clear-linux/tutorials/dlrs/dlrs.rst index ff3f2ca4..3098432c 100644 --- a/source/clear-linux/tutorials/dlrs/dlrs.rst +++ b/source/clear-linux/tutorials/dlrs/dlrs.rst @@ -32,7 +32,7 @@ The Deep Learning Reference Stack is available in the following versions: * `PyTorch with Intel MKL-DNN`_, which includes PyTorch optimized using Intel® Math Kernel Library (Intel® MKL) and Intel MKL-DNN. -.. note:: +.. important:: To take advantage of the Intel® AVX-512 and VNNI functionality with the Deep Learning Reference Stack, you must use the following hardware: @@ -185,6 +185,11 @@ Kubeflow multi-node benchmarks The benchmark workload runs in a Kubernetes cluster. The tutorial uses `Kubeflow`_ for the Machine Learning workload deployment on three nodes. +.. warning:: + + If you choose the Intel® MKL-DNN or Intel® MKL-DNN-VNNI image, your platform must support the Intel® AVX-512 instruction set. Otherwise, an *illegal instruction* error may appear, and you won’t be able to complete this tutorial. + + Kubernetes setup ================ @@ -236,9 +241,11 @@ or inject a python script or larger shell script into the container. Images ====== -You must add `launcher.py` to the Docker image to include the Deep +You must add `launcher.py`_ to the Docker image to include the Deep Learning Reference Stack and put the benchmarks repo in the correct -location. From the Docker image, run the following: +location. Note that this tutorial uses Kubeflow v0.4.0, and cannot guarantee results if you use a different version. + +From the Docker image, run the following: .. code-block:: bash @@ -486,6 +493,8 @@ Related topics .. _kubectl overview: https://kubernetes.io/docs/reference/kubectl/overview/ +.. _launcher.py: https://github.com/clearlinux/dockerfiles/tree/master/stacks/dlrs/kubeflow + .. _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