benchmarks
TensorFlow, TensorFlow-Lite Pytorch, Torchvision, TensorRT Benchmarks
MIT License
Stars
21
Ecosystems:
PyTorch,
Tensorflow
Benchmarks
About
Benchmarks such as TensorFlow, TensorFlow-Lite, PyTorch, Torchvision, etc...
Summary
HostPC / Google Colaboratory
| Description | Benchmarks | Framework | HW |
|---|---|---|---|
| TF2 Detection Models | AP, Latency | TensorFlowXLA Auto-clustering | Local PC GTX1070 |
| TF2 Detection Models | AP, Latency | TensorFlowTF-TRTXLA Auto-clustering | Google Colab Tesla P100 |
| TF2 Detection Models | AP, Latency | TensorFlow | Google Colab Tesla V100 |
| TF2 Detection Models | Latency | TensorFlow | V100 vs P100 vs GTX1070 |
| Torchvision detection models | AP, Latency, YouTube video | PyTorch, Torchvision | Local PC GTX1070 |
Raspberry Pi / Coral Dev Board + EdgeTPU
| Description | Benchmarks | Framework | HW |
|---|---|---|---|
| TensorFlow 1 Detection Model Zoo | AP, Latency | TensorFlow LiteXNNPACK delegateARM NN delegateEdgeTPU delegate | Raspberry Pi 2, 3, 4Coral USB Accelerator |
| EdgeTPU Models | Latency | TensorFlow LiteEdgeTPU delegate | Coral Dev Board |
| EfficientDet-Lite | AP, Latency | TensorFlow LitXNNPACK delegateARM NN delegateEdgeTPU delegate | Raspberry Pi 2, 3, 4Coral USB Accelerator |
| YOLOX TensorFlow LiteM | AP, Latency | TensorFlow LiteXNNPACK delegate | Raspberry Pi 4 |
| TF-Lite SpaghettiNet EdgeTpu Latency vs max_detections | AP, Latency | TensorFlow Lite EdgeTPU delegate | Raspberry Pi 4 + Coral USB AcceleratorCoral Dev Board |
Jetson
| Description | Benchmarks | Framework | HW |
|---|---|---|---|
| TensorRT EfficientDet and TensorFlow Object Detection API | Latency,YouTube video | TensorRT | Jetson Nano |
| TensorRT Autoseg-EdgeTPU andDeepLab v3+ MobilenetEdgeTPUV2 | Latency,YouTube video | TensorRT | Jetson Nano |
| TensorRT Fast-SCNN | Latency,YouTube video | TensorRT | Jetson Nano |
| TensorRT Ultra-Fast-Lane-Detection | Latency,YouTube video | TensorRT | Jetson Nano |
TensorFlow Lite on RISC-V SoC
| Description | Benchmarks | Framework | HW |
|---|---|---|---|
| Sipeed Lichee RV Dock Allwinner D1 | Latency | TensorFlow Lite | CPU |
| StarFive VisionFive 2 | Latency | TensorFlow Lite | CPU, GPU |
Reference
- TensorFlow 1 Detection Model Zoo
- TensorFlow 2 Detection Model Zoo
- TensorFlow XLA Auto-clustering
- TF-TRT (tensorflow / tensorrt)
- FP32 XNNPACK: PINTO0309/TensorflowLite-bin
- EfficientDet
- The Arm NN TensorFlow Lite delegate
- ARM Compute Librarry
- Coral EdgeTPU
- PINTO0309/Bazel_bin
- PyTorch
- Torchvision
- TensorRT
- meta-tensorflow-lite
- Yocto Project