Accueil Explorer Aide
Connexion
RONCC
/
NVIDIA-AI-Image-Classification
Suivre 1
Voter 0
Fork 0
Fichiers Tickets 0 Pull Requests 0 Wiki

Aucune description

Aborescence: 0aced03dc4
Branches Tags
NVIDIA-AI-Im...
ZIP TAR.GZ
John Welsh 0aced03dc4 initial commit il y a 7 ans
data 0aced03dc4 initial commit il y a 7 ans
examples 0aced03dc4 initial commit il y a 7 ans
scripts 0aced03dc4 initial commit il y a 7 ans
src 0aced03dc4 initial commit il y a 7 ans
CMakeLists.txt 0aced03dc4 initial commit il y a 7 ans
INSTALL.md 0aced03dc4 initial commit il y a 7 ans
LICENSE.md 0aced03dc4 initial commit il y a 7 ans
README.md 0aced03dc4 initial commit il y a 7 ans

README.md

TensorFlow->TensorRT Image Classification

This contains examples, scripts and code related to image classification using TensorFlow models (from here) converted to TensorRT. Converting TensorFlow models to TensorRT offers significant performance gains on the Jetson TX2 as seen below.

Quick Start

  1. Follow the installation guide.

  2. Download the pretrained TensorFlow models and example images.

    source scripts/download_models.sh
source scripts/download_images.sh

  3. Convert the pretrained models to frozen graphs.

    python scripts/models_to_frozen_graphs.py

  4. Convert the frozen graphs to optimized TensorRT engines.

    python scripts/frozen_graphs_to_plans.py

  5. Execute the Inception V1 model on a single image.

    ./build/examples/classify_image/classify_image data/images/gordon_setter.jpg data/plans/inception_v1.plan data/imagenet_labels_1001.txt input InceptionV1/Logits/SpatialSqueeze inception

For more details, read through the examples link.

Default Models

The table below shows various details related to the default models ported from the TensorFlow slim model zoo.

Model Input Size TensorRT (TX2 / Half) TensorRT (TX2 / Float) TensorFlow (TX2 / Float) Input Name Output Name Preprocessing Fn.
inception_v1 224x224 7.98ms 12.8ms 27.6ms input InceptionV1/Logits/SpatialSqueeze inception
inception_v3 299x299 26.3ms 46.1ms 98.4ms input InceptionV3/Logits/SpatialSqueeze inception
inception_v4 299x299 52.1ms 88.2ms 176ms input InceptionV4/Logits/Logits/BiasAdd inception
inception_resnet_v2 299x299 53.0ms 98.7ms 168ms input InceptionResnetV2/Logits/Logits/BiasAdd inception
resnet_v1_50 224x224 15.7ms 27.1ms 63.9ms input resnet_v1_50/SpatialSqueeze vgg
resnet_v1_101 224x224 29.9ms 51.8ms 107ms input resnet_v1_101/SpatialSqueeze vgg
resnet_v1_152 224x224 42.6ms 78.2ms 157ms input resnet_v1_152/SpatialSqueeze vgg
resnet_v2_50 299x299 27.5ms 44.4ms 92.2ms input resnet_v2_50/SpatialSqueeze inception
resnet_v2_101 299x299 49.2ms 83.1ms 160ms input resnet_v2_101/SpatialSqueeze inception
resnet_v2_152 299x299 74.6ms 124ms 230ms input resnet_v2_152/SpatialSqueeze inception
mobilenet_v1_0p25_128 128x128 2.67ms 2.65ms 15.7ms input MobilenetV1/Logits/SpatialSqueeze inception
mobilenet_v1_0p5_160 160x160 3.95ms 4.00ms 16.9ms input MobilenetV1/Logits/SpatialSqueeze inception
mobilenet_v1_1p0_224 224x224 12.9ms 12.9ms 24.4ms input MobilenetV1/Logits/SpatialSqueeze inception
vgg_16 224x224 38.2ms 79.2ms 171ms input vgg_16/fc8/BiasAdd vgg

The times recorded include data transfer to GPU, network execution, and data transfer back from GPU. Time does not include preprocessing. See scripts/test_tf.py, scripts/test_trt.py, and src/test/test_trt.cu for implementation details. To reproduce the timings run

python scripts/test_tf.py
python scripts/test_trt.py

The timing results will be located in data/test_output_tf.txt and data/test_output_trt.txt. Note that you must download and convert the models (as in the quick start) prior to running the benchmark scripts.