| --- |
| library_name: pytorch |
| license: mit |
| tags: |
| - bu_auto |
| - android |
| pipeline_tag: other |
|
|
| --- |
| |
|  |
|
|
| # StateTransformer: Optimized for Qualcomm Devices |
|
|
| StateTransformer is a transformer-based model designed for trajectory prediction in self-driving scenarios. It integrates rasterized map data, agent context, and temporal dynamics to generate accurate future trajectories. |
|
|
| This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.62.2/src/qai_hub_models/models/statetransformer) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary). |
|
|
| Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/statetransformer/releases/v0.62.2/statetransformer-onnx-float.zip) |
| | TFLITE | float | Universal | | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/statetransformer/releases/v0.62.2/statetransformer-tflite-float.zip) |
|
|
| For more device-specific assets and performance metrics, visit **[StateTransformer on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/statetransformer)**. |
|
|
|
|
| ### Option 2: Export with Custom Configurations |
|
|
| Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.62.2/src/qai_hub_models/models/statetransformer) 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 [StateTransformer on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.62.2/src/qai_hub_models/models/statetransformer) for usage instructions. |
|
|
| ## Model Details |
|
|
| **Model Type:** Model_use_case.driver_assistance |
| |
| **Model Stats:** |
| - Input resolution: 1x224x224x58, 1x224x224x58, 1x4x7 |
| - Model checkpoint: pretrained-mixtral-small |
| - Model size (float): 348 MB |
| - Number of parameters: 90.7M |
| |
| ## Performance Summary |
| | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
| |---|---|---|---|---|---|--- |
| | StateTransformer | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 291.136 ms | 200 - 5703 MB | NPU |
| | StateTransformer | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 349.435 ms | 100 - 142 MB | NPU |
| | StateTransformer | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 373.184 ms | 196 - 200 MB | NPU |
| | StateTransformer | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 315.741 ms | 106 - 147 MB | NPU |
| | StateTransformer | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 236.364 ms | 168 - 5681 MB | NPU |
| | StateTransformer | ONNX | float | Snapdragon® 8 Elite Mobile | 236.364 ms | 168 - 5681 MB | NPU |
| | StateTransformer | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 212.445 ms | 183 - 5749 MB | NPU |
| | StateTransformer | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1528.046 ms | 294 - 816 MB | CPU |
| | StateTransformer | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1115.87 ms | 241 - 394 MB | CPU |
| | StateTransformer | TFLITE | float | Qualcomm® SA8650P | 792.928 ms | 452 - 465 MB | CPU |
| | StateTransformer | TFLITE | float | Qualcomm® SA8255P | 792.928 ms | 452 - 465 MB | CPU |
| | StateTransformer | TFLITE | float | Qualcomm® QCS8450 | 808.833 ms | 450 - 470 MB | CPU |
| | StateTransformer | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1309.723 ms | 293 - 815 MB | CPU |
| | StateTransformer | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 645.781 ms | 267 - 4157 MB | CPU |
| | StateTransformer | TFLITE | float | Snapdragon® 8 Elite Mobile | 645.781 ms | 267 - 4157 MB | CPU |
| |
| ## License |
| * The license for the original implementation of StateTransformer can be found |
| [here](https://github.com/Tsinghua-MARS-Lab/StateTransformer/blob/main/setup.py). |
| |
| |
| |
| ## Community |
| * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI. |
| * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com). |
| |
