Deep Layer Aggregation
Paper • 1707.06484 • Published
DLA-102-X: Optimized for Qualcomm Devices
dla is a machine learning model that can classify images from the Imagenet dataset. It is a architecture which can be used as backbone.
This is based on the implementation of DLA-102-X 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 2.42, ONNX Runtime 1.24.1 | Download |
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| QNN_DLC | float | Universal | QAIRT 2.43 | Download |
| QNN_DLC | w8a8 | Universal | QAIRT 2.43 | Download |
| TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | Download |
| TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | Download |
For more device-specific assets and performance metrics, visit DLA-102-X 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 DLA-102-X on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.image_classification
Model Stats:
- Model checkpoint: Imagenet
- Input resolution: 224x224
- Number of parameters: 26.3M
- Model size (float): 100 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| DLA-102-X | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.191 ms | 1 - 108 MB | NPU |
| DLA-102-X | ONNX | float | Snapdragon® X2 Elite | 1.332 ms | 54 - 54 MB | NPU |
| DLA-102-X | ONNX | float | Snapdragon® X Elite | 2.921 ms | 54 - 54 MB | NPU |
| DLA-102-X | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.88 ms | 0 - 168 MB | NPU |
| DLA-102-X | ONNX | float | Qualcomm® QCS8550 (Proxy) | 2.65 ms | 0 - 60 MB | NPU |
| DLA-102-X | ONNX | float | Qualcomm® QCS9075 | 3.972 ms | 1 - 4 MB | NPU |
| DLA-102-X | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.556 ms | 0 - 105 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.733 ms | 0 - 119 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Snapdragon® X2 Elite | 0.726 ms | 28 - 28 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Snapdragon® X Elite | 1.626 ms | 27 - 27 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.015 ms | 0 - 150 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Qualcomm® QCS6490 | 74.814 ms | 20 - 37 MB | CPU |
| DLA-102-X | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.403 ms | 0 - 31 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Qualcomm® QCS9075 | 1.629 ms | 0 - 3 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Qualcomm® QCM6690 | 40.918 ms | 5 - 17 MB | CPU |
| DLA-102-X | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.881 ms | 0 - 117 MB | NPU |
| DLA-102-X | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 38.757 ms | 17 - 28 MB | CPU |
| DLA-102-X | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.229 ms | 1 - 106 MB | NPU |
| DLA-102-X | QNN_DLC | float | Snapdragon® X2 Elite | 1.679 ms | 1 - 1 MB | NPU |
| DLA-102-X | QNN_DLC | float | Snapdragon® X Elite | 3.176 ms | 1 - 1 MB | NPU |
| DLA-102-X | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 2.015 ms | 0 - 150 MB | NPU |
| DLA-102-X | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 14.36 ms | 1 - 102 MB | NPU |
| DLA-102-X | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 2.874 ms | 1 - 3 MB | NPU |
| DLA-102-X | QNN_DLC | float | Qualcomm® QCS9075 | 4.238 ms | 1 - 3 MB | NPU |
| DLA-102-X | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 4.572 ms | 0 - 125 MB | NPU |
| DLA-102-X | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.605 ms | 1 - 104 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.663 ms | 0 - 107 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.924 ms | 0 - 0 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.589 ms | 0 - 0 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.983 ms | 0 - 131 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 3.874 ms | 0 - 2 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 3.253 ms | 0 - 103 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.397 ms | 0 - 2 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 1.604 ms | 0 - 2 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 10.63 ms | 0 - 229 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 1.857 ms | 0 - 131 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.792 ms | 0 - 105 MB | NPU |
| DLA-102-X | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.727 ms | 0 - 101 MB | NPU |
| DLA-102-X | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.214 ms | 0 - 164 MB | NPU |
| DLA-102-X | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.964 ms | 0 - 209 MB | NPU |
| DLA-102-X | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 14.38 ms | 0 - 162 MB | NPU |
| DLA-102-X | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 2.765 ms | 0 - 3 MB | NPU |
| DLA-102-X | TFLITE | float | Qualcomm® QCS9075 | 4.256 ms | 0 - 57 MB | NPU |
| DLA-102-X | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 4.477 ms | 0 - 179 MB | NPU |
| DLA-102-X | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.585 ms | 0 - 157 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.591 ms | 0 - 102 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.843 ms | 0 - 131 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Qualcomm® QCS6490 | 3.616 ms | 0 - 29 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 3.0 ms | 0 - 99 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.128 ms | 0 - 2 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Qualcomm® QCS9075 | 1.37 ms | 0 - 29 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Qualcomm® QCM6690 | 10.069 ms | 0 - 208 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 1.655 ms | 0 - 136 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.709 ms | 0 - 92 MB | NPU |
| DLA-102-X | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.491 ms | 0 - 92 MB | NPU |
License
- The license for the original implementation of DLA-102-X 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.