Learn how to bring up ExecuTorch executor_runner firmware on the NXP FRDM i.MX 93 Cortex-M33 using Linux RemoteProc, compile .pte models for Ethos-U65, and run inference with NPU acceleration.
Learn how to train a CNN image classification model using PyTorch, convert it to ExecuTorch format, and run it as an interactive mini-game on Arm-based edge devices.
Learn how to identify Arm-based targets for TinyML, install Fixed Virtual Platforms, deploy ExecuTorch models on Corstone-320 FVP, and visualize model execution using the FVP graphical interface.
Deploy a MobileNetV2 image classification model to an Alif Ensemble E8 DevKit and run inference on the Ethos-U85 NPU.
Evaluation Conditions
Image Classification · Alif DK-E8 · INT8 · 299×299 · ImageNet-1k (val)
| Metric | Resultsvs baseline | Change | |
|---|---|---|---|
| End to end latency p50 | 237.13ms-ms | ||
| End to end latency p90 | 237.13ms- |
| Peak memory | 7.48MB-MB |
| Frames per second | 4.22fps-fps |
| Accuracy - Top 1 | 76.45%77.30% | −0.85% |
| Accuracy - Top 5 | 93.01%93.45% | −0.44% |
| Configuration | Baseline | Arm Optimized |
|---|---|---|
| Dataset | ImageNet-1k (val) | ImageNet-1k (val) |
| Sample count | 5,000 | 5,000 |
| Batch size | 1 | 1 |
| Input resolution |
The Optimized Model is provided as a reference implementation to demonstrate and evaluate execution and performance on Arm-based systems. It is not a production-ready or supported solution.
Arm’s publication of the Optimized Model does not constitute an endorsement or certification of the Original Model or a representation that the Optimized Model is suitable for production use or any particular purpose.
To the fullest extent permitted by applicable law (i) the Optimized Model is provided “as is.” Arm makes no representations or warranties that the Original Model, the Optimized Model or their outputs are accurate, safe, secure, non-infringing, legally compliant, suitable for production use or fit for any particular purpose; and (ii) Arm will not be liable for any loss or damage arising from or in connection with the Optimized Model, its use or its outputs.
You are responsible for independently evaluating the Optimized Model, its outputs and its suitability for your intended use, including compliance with applicable legal, regulatory, safety and security requirements.
Arm does not commit to provide ongoing support, maintenance or updates for the Optimized Model. Any use of or reliance on the Optimized Model or its outputs is at your own risk.
Use of the Portal is subject to the Arm Website Terms and Conditions of Use.
| 299x299 |
299x299 |
| Runs | 1 | 100 |
| Warmup runs | N/A | 10 |