EfficientNetV2: Smaller Models and Faster Training
Paper • 2104.00298 • Published • 1
EfficientNet-V2-s: Optimized for Qualcomm Devices
EfficientNetV2-s is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.
This is based on the implementation of EfficientNet-V2-s found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.
Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.
Getting Started
There are two ways to deploy this model on your device:
Option 1: Download Pre-Exported Models
Below are pre-exported model assets ready for deployment.
| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| ONNX | float | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | Download |
| ONNX | w8a16 | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | Download |
| QNN_DLC | float | Universal | QAIRT 2.50 | Download |
| QNN_DLC | w8a16 | Universal | QAIRT 2.50 | Download |
| TFLITE | float | Universal | QAIRT 2.50 | Download |
For more device-specific assets and performance metrics, visit EfficientNet-V2-s on Qualcomm® AI Hub.
Option 2: Export with Custom Configurations
Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations
This option is ideal if you need to customize the model beyond the default configuration provided here.
See our repository for EfficientNet-V2-s on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.image_classification
Model Stats:
- Input resolution: 384x384
- Model checkpoint: Imagenet
- Model size (float): 81.7 MB
- Model size (w8a16): 27.2 MB
- Number of parameters: 21.4M
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| EfficientNet-V2-s | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.254 ms | 1 - 199 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.066 ms | 0 - 194 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Snapdragon® X2 Elite | 2.961 ms | 2 - 2 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Snapdragon® X Elite | 5.583 ms | 47 - 47 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.076 ms | 0 - 169 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 11.186 ms | 2 - 197 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 8.095 ms | 2 - 7 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.307 ms | 0 - 48 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Qualcomm® QCS8450 | 11.186 ms | 2 - 197 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 7.627 ms | 1 - 6 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 5.583 ms | 47 - 47 MB | NPU |
| EfficientNet-V2-s | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 3.066 ms | 0 - 194 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.992 ms | 0 - 164 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.521 ms | 0 - 163 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Snapdragon® X2 Elite | 2.328 ms | 2 - 2 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Snapdragon® X Elite | 5.705 ms | 24 - 24 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.648 ms | 0 - 209 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 6.733 ms | 1 - 218 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 16.228 ms | 1 - 4 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 5.136 ms | 1 - 5 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.44 ms | 0 - 29 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® QCS8450 | 6.733 ms | 1 - 218 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 5.734 ms | 1 - 4 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 5.705 ms | 24 - 24 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 37.136 ms | 1 - 265 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 6.59 ms | 1 - 265 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.521 ms | 0 - 163 MB | NPU |
| EfficientNet-V2-s | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 6.59 ms | 1 - 265 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.336 ms | 2 - 82 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.16 ms | 0 - 77 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Snapdragon® X2 Elite | 3.405 ms | 2 - 2 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Snapdragon® X Elite | 6.284 ms | 2 - 2 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 4.374 ms | 2 - 159 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 13.07 ms | 0 - 188 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 8.181 ms | 2 - 6 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.719 ms | 2 - 4 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® SA8775P | 8.198 ms | 2 - 76 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® SA8650P | 8.198 ms | 2 - 76 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® SA8255P | 8.198 ms | 2 - 76 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® QCS8450 | 13.07 ms | 0 - 188 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 7.955 ms | 4 - 7 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 6.284 ms | 2 - 2 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 3.16 ms | 0 - 77 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | float | Qualcomm® SA8295P | 12.913 ms | 2 - 108 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.216 ms | 1 - 153 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.82 ms | 0 - 148 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 2.864 ms | 1 - 1 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Snapdragon® X Elite | 6.802 ms | 1 - 1 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 4.213 ms | 0 - 183 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Snapdragon® 8 Gen 1 Mobile | 7.866 ms | 1 - 194 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 18.877 ms | 1 - 4 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 5.793 ms | 1 - 5 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 6.178 ms | 1 - 2 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® SA8650P | 6.716 ms | 1 - 136 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® SA8255P | 6.716 ms | 1 - 136 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® QCS8450 | 7.866 ms | 1 - 194 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 6.45 ms | 1 - 4 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 6.802 ms | 1 - 1 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 40.11 ms | 1 - 251 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 8.078 ms | 1 - 253 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.82 ms | 0 - 148 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Qualcomm® SA8295P | 8.043 ms | 1 - 141 MB | NPU |
| EfficientNet-V2-s | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 8.078 ms | 1 - 253 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.332 ms | 0 - 121 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.157 ms | 0 - 117 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 4.373 ms | 0 - 202 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 13.267 ms | 0 - 232 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 8.19 ms | 0 - 52 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.724 ms | 0 - 2 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® SA8775P | 8.169 ms | 0 - 120 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® SA8650P | 8.169 ms | 0 - 120 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® SA8255P | 8.169 ms | 0 - 120 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® QCS8450 | 13.267 ms | 0 - 232 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 7.933 ms | 0 - 51 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 3.157 ms | 0 - 117 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® SA7255P | 25.757 ms | 0 - 117 MB | NPU |
| EfficientNet-V2-s | TFLITE | float | Qualcomm® SA8295P | 12.866 ms | 0 - 149 MB | NPU |
License
- The license for the original implementation of EfficientNet-V2-s can be found here.
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.