English|中文

ResNext50 Inference for Tensorflow

This repository provides a script and recipe to Inference the ResNext50 model.

Notice

This sample only provides reference for you to learn the Ascend software stack and is not for commercial purposes.

Before starting, please pay attention to the following adaptation conditions. If they do not match, may leading in failure.

Quick Start Guide

1. Clone the respository

git clone https://gitee.com/ascend/ModelZoo-TensorFlow.git
cd Modelzoo-TensorFlow/ACL_TensorFlow/built-in/cv/ResNext50_for_ACL

2. Download and preprocess the dataset

1. Download the ImageNet2012 dataset by yourself

3. Offline Inference

Convert pb to om.

• configure the env

  Please follow the guide to set the envs

• convert pb to om

  pb download link

  For Ascend310:

  atc --model=resnext50_tf.pb --framework=3 --output=resnext50_tf_aipp --output_type=FP32 --soc_version=Ascend310 --input_shape="input:1,224,224,3" --log=info --insert_op_conf=resnext50_tf_aipp.cfg
  

  For Ascend310P3:

  atc --model=resnext50_tf.pb --framework=3 --output=resnext50_tf_aipp --output_type=FP32 --soc_version=Ascend310P3 --input_shape="input:1,224,224,3" --log=info --insert_op_conf=resnext50_tf_aipp.cfg
  

• Build the program

  For Ascend310:

  unset ASCEND310P3_DVPP
  bash build.sh
  

  For Ascend310P3:

  export ASCEND310P3_DVPP=1
  bash build.sh
  

• Run the program:

  cd scripts
  bash benchmark_tf.sh --batchSize=1 --modelType=resnext50 --imgType=raw --precision=fp16 --outputType=fp32 --useDvpp=1 --deviceId=0 --modelPath=resnext50_tf_aipp.om --dataPath=image-1024 --trueValuePath=val_lable.txt
  

Performance

Result

Our result were obtained by running the applicable inference script. To achieve the same results, follow the steps in the Quick Start Guide.

Inference accuracy results