v0.58.0
Browse filesSee https://github.com/qualcomm/ai-hub-models/releases/v0.58.0 for changelog.
- README.md +70 -53
- release_assets.json +6 -6
README.md
CHANGED
|
@@ -15,7 +15,7 @@ pipeline_tag: object-detection
|
|
| 15 |
DETR is a machine learning model that can detect objects (trained on COCO dataset).
|
| 16 |
|
| 17 |
This is based on the implementation of DETR-ResNet50 found [here](https://github.com/facebookresearch/detr).
|
| 18 |
-
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.
|
| 19 |
|
| 20 |
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.
|
| 21 |
|
|
@@ -28,25 +28,25 @@ Below are pre-exported model assets ready for deployment.
|
|
| 28 |
|
| 29 |
| Runtime | Precision | Chipset | SDK Versions | Download |
|
| 30 |
|---|---|---|---|---|
|
| 31 |
-
| ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 32 |
-
| ONNX | w8a16_mixed_fp16 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 33 |
-
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 34 |
-
| QNN_DLC | w8a16_mixed_fp16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 35 |
-
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 36 |
|
| 37 |
For more device-specific assets and performance metrics, visit **[DETR-ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/detr_resnet50)**.
|
| 38 |
|
| 39 |
|
| 40 |
### Option 2: Export with Custom Configurations
|
| 41 |
|
| 42 |
-
Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 43 |
- Custom weights (e.g., fine-tuned checkpoints)
|
| 44 |
- Custom input shapes
|
| 45 |
- Target device and runtime configurations
|
| 46 |
|
| 47 |
This option is ideal if you need to customize the model beyond the default configuration provided here.
|
| 48 |
|
| 49 |
-
See our repository for [DETR-ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 50 |
|
| 51 |
## Model Details
|
| 52 |
|
|
@@ -61,51 +61,68 @@ See our repository for [DETR-ResNet50 on GitHub](https://github.com/qualcomm/ai-
|
|
| 61 |
## Performance Summary
|
| 62 |
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|
| 63 |
|---|---|---|---|---|---|---
|
| 64 |
-
| DETR-ResNet50 | ONNX | float | Snapdragon® X2 Elite | 8.
|
| 65 |
-
| DETR-ResNet50 | ONNX | float | Snapdragon® X Elite | 17.
|
| 66 |
-
| DETR-ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 13.
|
| 67 |
-
| DETR-ResNet50 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 36.
|
| 68 |
-
| DETR-ResNet50 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 17.
|
| 69 |
-
| DETR-ResNet50 | ONNX | float | Qualcomm® QCS8450 | 36.
|
| 70 |
-
| DETR-ResNet50 | ONNX | float |
|
| 71 |
-
| DETR-ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.
|
| 72 |
-
| DETR-ResNet50 | ONNX | float |
|
| 73 |
-
| DETR-ResNet50 | ONNX | float | Qualcomm®
|
| 74 |
-
| DETR-ResNet50 | ONNX | float | Qualcomm®
|
| 75 |
-
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 52.
|
| 76 |
-
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) |
|
| 77 |
-
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Mobile | 41.
|
| 78 |
-
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 |
|
| 79 |
-
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 |
|
| 80 |
-
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Qualcomm®
|
| 81 |
-
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite |
|
| 82 |
-
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 19.
|
| 83 |
-
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 13.
|
| 84 |
-
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile |
|
| 85 |
-
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8275 |
|
| 86 |
-
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 19.
|
| 87 |
-
| DETR-ResNet50 | QNN_DLC | float | Qualcomm®
|
| 88 |
-
| DETR-ResNet50 | QNN_DLC | float |
|
| 89 |
-
| DETR-ResNet50 | QNN_DLC | float | Qualcomm®
|
| 90 |
-
| DETR-ResNet50 | QNN_DLC | float |
|
| 91 |
-
| DETR-ResNet50 | QNN_DLC | float | Qualcomm®
|
| 92 |
-
| DETR-ResNet50 | QNN_DLC | float |
|
| 93 |
-
| DETR-ResNet50 | QNN_DLC | float | Qualcomm®
|
| 94 |
-
| DETR-ResNet50 | QNN_DLC | float |
|
| 95 |
-
| DETR-ResNet50 |
|
| 96 |
-
| DETR-ResNet50 |
|
| 97 |
-
| DETR-ResNet50 |
|
| 98 |
-
| DETR-ResNet50 |
|
| 99 |
-
| DETR-ResNet50 |
|
| 100 |
-
| DETR-ResNet50 |
|
| 101 |
-
| DETR-ResNet50 |
|
| 102 |
-
| DETR-ResNet50 |
|
| 103 |
-
| DETR-ResNet50 |
|
| 104 |
-
| DETR-ResNet50 |
|
| 105 |
-
| DETR-ResNet50 |
|
| 106 |
-
| DETR-ResNet50 |
|
| 107 |
-
| DETR-ResNet50 |
|
| 108 |
-
| DETR-ResNet50 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 109 |
|
| 110 |
## License
|
| 111 |
* The license for the original implementation of DETR-ResNet50 can be found
|
|
|
|
| 15 |
DETR is a machine learning model that can detect objects (trained on COCO dataset).
|
| 16 |
|
| 17 |
This is based on the implementation of DETR-ResNet50 found [here](https://github.com/facebookresearch/detr).
|
| 18 |
+
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/v0.58.0/src/qai_hub_models/models/detr_resnet50) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
|
| 19 |
|
| 20 |
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.
|
| 21 |
|
|
|
|
| 28 |
|
| 29 |
| Runtime | Precision | Chipset | SDK Versions | Download |
|
| 30 |
|---|---|---|---|---|
|
| 31 |
+
| ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-onnx-float.zip)
|
| 32 |
+
| ONNX | w8a16_mixed_fp16 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-onnx-w8a16_mixed_fp16.zip)
|
| 33 |
+
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-qnn_dlc-float.zip)
|
| 34 |
+
| QNN_DLC | w8a16_mixed_fp16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-qnn_dlc-w8a16_mixed_fp16.zip)
|
| 35 |
+
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-tflite-float.zip)
|
| 36 |
|
| 37 |
For more device-specific assets and performance metrics, visit **[DETR-ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/detr_resnet50)**.
|
| 38 |
|
| 39 |
|
| 40 |
### Option 2: Export with Custom Configurations
|
| 41 |
|
| 42 |
+
Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/detr_resnet50) Python library to compile and export the model with your own:
|
| 43 |
- Custom weights (e.g., fine-tuned checkpoints)
|
| 44 |
- Custom input shapes
|
| 45 |
- Target device and runtime configurations
|
| 46 |
|
| 47 |
This option is ideal if you need to customize the model beyond the default configuration provided here.
|
| 48 |
|
| 49 |
+
See our repository for [DETR-ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/detr_resnet50) for usage instructions.
|
| 50 |
|
| 51 |
## Model Details
|
| 52 |
|
|
|
|
| 61 |
## Performance Summary
|
| 62 |
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|
| 63 |
|---|---|---|---|---|---|---
|
| 64 |
+
| DETR-ResNet50 | ONNX | float | Snapdragon® X2 Elite | 8.209 ms | 5 - 5 MB | NPU
|
| 65 |
+
| DETR-ResNet50 | ONNX | float | Snapdragon® X Elite | 17.912 ms | 78 - 78 MB | NPU
|
| 66 |
+
| DETR-ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 13.422 ms | 5 - 375 MB | NPU
|
| 67 |
+
| DETR-ResNet50 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 36.141 ms | 3 - 320 MB | NPU
|
| 68 |
+
| DETR-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 17.805 ms | 0 - 95 MB | NPU
|
| 69 |
+
| DETR-ResNet50 | ONNX | float | Qualcomm® QCS8450 | 36.141 ms | 3 - 320 MB | NPU
|
| 70 |
+
| DETR-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 27.089 ms | 5 - 12 MB | NPU
|
| 71 |
+
| DETR-ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.44 ms | 3 - 318 MB | NPU
|
| 72 |
+
| DETR-ResNet50 | ONNX | float | Snapdragon® 8 Elite Mobile | 9.953 ms | 1 - 270 MB | NPU
|
| 73 |
+
| DETR-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 9.953 ms | 1 - 270 MB | NPU
|
| 74 |
+
| DETR-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 17.912 ms | 78 - 78 MB | NPU
|
| 75 |
+
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 52.086 ms | 19 - 443 MB | NPU
|
| 76 |
+
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 68.763 ms | 19 - 22 MB | NPU
|
| 77 |
+
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Mobile | 41.718 ms | 21 - 369 MB | NPU
|
| 78 |
+
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 75.245 ms | 22 - 24 MB | NPU
|
| 79 |
+
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 33.083 ms | 21 - 391 MB | NPU
|
| 80 |
+
| DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 41.718 ms | 21 - 369 MB | NPU
|
| 81 |
+
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 8.974 ms | 5 - 5 MB | NPU
|
| 82 |
+
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 19.529 ms | 5 - 5 MB | NPU
|
| 83 |
+
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 13.711 ms | 5 - 361 MB | NPU
|
| 84 |
+
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 41.438 ms | 5 - 302 MB | NPU
|
| 85 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8275 | 85.977 ms | 1 - 267 MB | NPU
|
| 86 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 19.171 ms | 5 - 284 MB | NPU
|
| 87 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 26.844 ms | 1 - 300 MB | NPU
|
| 88 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8650P | 26.844 ms | 1 - 300 MB | NPU
|
| 89 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8255P | 26.844 ms | 1 - 300 MB | NPU
|
| 90 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 41.438 ms | 5 - 302 MB | NPU
|
| 91 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 27.448 ms | 5 - 11 MB | NPU
|
| 92 |
+
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.912 ms | 1 - 286 MB | NPU
|
| 93 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 85.977 ms | 1 - 267 MB | NPU
|
| 94 |
+
| DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 10.466 ms | 0 - 275 MB | NPU
|
| 95 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 31.183 ms | 0 - 224 MB | NPU
|
| 96 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 10.466 ms | 0 - 275 MB | NPU
|
| 97 |
+
| DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 19.529 ms | 5 - 5 MB | NPU
|
| 98 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 18.306 ms | 2 - 2 MB | NPU
|
| 99 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X Elite | 32.651 ms | 2 - 2 MB | NPU
|
| 100 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 24.626 ms | 2 - 445 MB | NPU
|
| 101 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® QCS8275 | 93.975 ms | 0 - 348 MB | NPU
|
| 102 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 33.735 ms | 3 - 417 MB | NPU
|
| 103 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8775P | 38.478 ms | 1 - 373 MB | NPU
|
| 104 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8650P | 38.478 ms | 1 - 373 MB | NPU
|
| 105 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8255P | 38.478 ms | 1 - 373 MB | NPU
|
| 106 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite Mobile | 19.41 ms | 0 - 357 MB | NPU
|
| 107 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 38.908 ms | 1 - 5 MB | NPU
|
| 108 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 16.934 ms | 2 - 401 MB | NPU
|
| 109 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA7255P | 93.975 ms | 0 - 348 MB | NPU
|
| 110 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 19.41 ms | 0 - 357 MB | NPU
|
| 111 |
+
| DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-X7181 | 32.651 ms | 2 - 2 MB | NPU
|
| 112 |
+
| DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 13.697 ms | 0 - 385 MB | NPU
|
| 113 |
+
| DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 41.515 ms | 0 - 326 MB | NPU
|
| 114 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8275 | 85.991 ms | 0 - 293 MB | NPU
|
| 115 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 19.367 ms | 0 - 3 MB | NPU
|
| 116 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® SA8775P | 26.916 ms | 0 - 324 MB | NPU
|
| 117 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® SA8650P | 26.916 ms | 0 - 324 MB | NPU
|
| 118 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® SA8255P | 26.916 ms | 0 - 324 MB | NPU
|
| 119 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8450 | 41.515 ms | 0 - 326 MB | NPU
|
| 120 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 27.086 ms | 0 - 88 MB | NPU
|
| 121 |
+
| DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.064 ms | 0 - 311 MB | NPU
|
| 122 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® SA7255P | 85.991 ms | 0 - 293 MB | NPU
|
| 123 |
+
| DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Mobile | 10.503 ms | 0 - 306 MB | NPU
|
| 124 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® SA8295P | 31.042 ms | 0 - 248 MB | NPU
|
| 125 |
+
| DETR-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 10.503 ms | 0 - 306 MB | NPU
|
| 126 |
|
| 127 |
## License
|
| 128 |
* The license for the original implementation of DETR-ResNet50 can be found
|
release_assets.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"version": "0.
|
| 3 |
"precisions": {
|
| 4 |
"float": {
|
| 5 |
"universal_assets": {
|
|
@@ -8,20 +8,20 @@
|
|
| 8 |
"qairt": "2.45.0.260326154327",
|
| 9 |
"litert": "1.4.4"
|
| 10 |
},
|
| 11 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 12 |
},
|
| 13 |
"qnn_dlc": {
|
| 14 |
"tool_versions": {
|
| 15 |
"qairt": "2.45.0.260326154327"
|
| 16 |
},
|
| 17 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 18 |
},
|
| 19 |
"onnx": {
|
| 20 |
"tool_versions": {
|
| 21 |
"qairt": "2.45.0.260326154327",
|
| 22 |
"onnx_runtime": "1.25.0"
|
| 23 |
},
|
| 24 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 25 |
}
|
| 26 |
}
|
| 27 |
},
|
|
@@ -31,14 +31,14 @@
|
|
| 31 |
"tool_versions": {
|
| 32 |
"qairt": "2.45.0.260326154327"
|
| 33 |
},
|
| 34 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 35 |
},
|
| 36 |
"onnx": {
|
| 37 |
"tool_versions": {
|
| 38 |
"qairt": "2.45.0.260326154327",
|
| 39 |
"onnx_runtime": "1.25.0"
|
| 40 |
},
|
| 41 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.
|
| 42 |
}
|
| 43 |
}
|
| 44 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"version": "0.58.0",
|
| 3 |
"precisions": {
|
| 4 |
"float": {
|
| 5 |
"universal_assets": {
|
|
|
|
| 8 |
"qairt": "2.45.0.260326154327",
|
| 9 |
"litert": "1.4.4"
|
| 10 |
},
|
| 11 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-tflite-float.zip"
|
| 12 |
},
|
| 13 |
"qnn_dlc": {
|
| 14 |
"tool_versions": {
|
| 15 |
"qairt": "2.45.0.260326154327"
|
| 16 |
},
|
| 17 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-qnn_dlc-float.zip"
|
| 18 |
},
|
| 19 |
"onnx": {
|
| 20 |
"tool_versions": {
|
| 21 |
"qairt": "2.45.0.260326154327",
|
| 22 |
"onnx_runtime": "1.25.0"
|
| 23 |
},
|
| 24 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-onnx-float.zip"
|
| 25 |
}
|
| 26 |
}
|
| 27 |
},
|
|
|
|
| 31 |
"tool_versions": {
|
| 32 |
"qairt": "2.45.0.260326154327"
|
| 33 |
},
|
| 34 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-qnn_dlc-w8a16_mixed_fp16.zip"
|
| 35 |
},
|
| 36 |
"onnx": {
|
| 37 |
"tool_versions": {
|
| 38 |
"qairt": "2.45.0.260326154327",
|
| 39 |
"onnx_runtime": "1.25.0"
|
| 40 |
},
|
| 41 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet50/releases/v0.58.0/detr_resnet50-onnx-w8a16_mixed_fp16.zip"
|
| 42 |
}
|
| 43 |
}
|
| 44 |
}
|