Description
This document provides links to step-by-step instructions on how to leverage reference model docker containers to run optimized open-source Deep Learning inference workloads using Intel® Extension for PyTorch* and Intel® Extension for TensorFlow* on the Intel® Data Center GPU Flex Series.
Base Containers
| AI Framework | Extension | Documentation |
|---|---|---|
| PyTorch | Intel® Extension for PyTorch* | Intel® Extension for PyTorch Container |
| TensorFlow | Intel® Extension for TensorFlow* | Intel® Extension for TensorFlow Container |
Optimized Workloads
The table below provides links to run each workload in a docker container. The containers were validated on a host running Linux*.
| Model | Framework | Mode and Documentation |
|---|---|---|
| DistilBERT | PyTorch | FP16 and FP32 Inference |
| DLRM v1 | PyTorch | FP16 Inference |
| EfficientNet B0,B3,B4 | TensorFlow | FP16 Inference |
| EfficientNet | PyTorch | FP32 FP16 BF16 Inference |
| FastPitch | PyTorch | FP16 Inference |
| Mask R-CNN | TensorFlow | FP16 Inference |
| ResNet50 v1.5 | PyTorch | INT8 Inference |
| ResNet50 v1.5 | TensorFlow | INT8 Inference |
| SSD-MobileNet v1 | TensorFlow | INT8 Inference |
| Stable Diffusion | PyTorch | FP16 Inference |
| Stable Diffusion | TensorFlow | FP32,FP16 Inference |
| UNet++ | PyTorch | FP16 Inference |
| Swin Transformer | PyTorch | FP16 Inference |
| Wide and Deep | TensorFlow | FP16 Inference |
| YOLO v5 | PyTorch | FP16 Inference |
Note:
- SSD-MobileNet v1 model is supported on older Intel® Extension for TensorFlow* v2.12 and Intel® Extension for PyTorch* 1.13.120+xpu versions.
- The other models in the list are validated on Intel® Extension for TensorFlow* v2.14 and Intel® Extension for PyTorch* 2.1.10+xpu versions.