From e1f996b913b6bf05173efe0ffdaa67d84a515595 Mon Sep 17 00:00:00 2001 From: Qi Zheng Date: Wed, 25 Dec 2019 01:27:16 +0000 Subject: [PATCH] Update Openvino README.md Signed-off-by: Qi Zheng --- openvino/README.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/openvino/README.md b/openvino/README.md index 918decb..87de6f1 100644 --- a/openvino/README.md +++ b/openvino/README.md @@ -59,9 +59,11 @@ The model precision to be chosen, FP32, FP16 or INT8. #### MODEL_PATH The chosen model path, automatically set by the entrypoint script. -For example, if using mobilenetv2-int8-tf-0001 model to do classification, two environment variables -need to be passed to the container. Details can refer to the deployment below. +For example, if using [`mobilenetv2-int8-tf-0001`](https://github.com/opencv/open_model_zoo/blob/master/models/intel/mobilenetv2-int8-tf-0001/description/mobilenetv2-int8-tf-0001.md) model to do classification, two environment variables need to be passed to the container. +Details can refer to the deployment below. +Note, use trained and quantized INT8 fixed-point precision model such as `mobilenetv2-int8-tf-0001` +on AVX512 VNNI platform could get [big performance advantage](https://www.intel.ai/vnni-enables-inference/). ## Deployment: