Hugging Face's logo

qualcomm
/
Conditional-DETR-ResNet50

Object Detection
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
34017ea
·
verified ·
1 Parent(s): 441dddf

v0.48.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.48.0 for changelog.

Files changed (1) hide show
  1. README.md +37 -37
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: object-detection
14
  DETR is a machine learning model that can detect objects (trained on COCO dataset).
15
 
16
  This is based on the implementation of Conditional-DETR-ResNet50 found [here](https://github.com/huggingface/transformers/tree/main/src/transformers/models/conditional_detr).
17
- This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/conditional_detr_resnet50) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  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.
20
 
@@ -27,25 +27,25 @@ Below are pre-exported model assets ready for deployment.
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
- | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.47.0/conditional_detr_resnet50-onnx-float.zip)
31
- | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.47.0/conditional_detr_resnet50-onnx-w8a16.zip)
32
- | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.47.0/conditional_detr_resnet50-qnn_dlc-float.zip)
33
- | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.47.0/conditional_detr_resnet50-qnn_dlc-w8a16.zip)
34
- | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.47.0/conditional_detr_resnet50-tflite-float.zip)
35
 
36
  For more device-specific assets and performance metrics, visit **[Conditional-DETR-ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/conditional_detr_resnet50)**.
37
 
38
 
39
  ### Option 2: Export with Custom Configurations
40
 
41
- Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/conditional_detr_resnet50) Python library to compile and export the model with your own:
42
  - Custom weights (e.g., fine-tuned checkpoints)
43
  - Custom input shapes
44
  - Target device and runtime configurations
45
 
46
  This option is ideal if you need to customize the model beyond the default configuration provided here.
47
 
48
- See our repository for [Conditional-DETR-ResNet50 on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/conditional_detr_resnet50) for usage instructions.
49
 
50
  ## Model Details
51
 
@@ -60,35 +60,35 @@ See our repository for [Conditional-DETR-ResNet50 on GitHub](https://github.com/
60
  ## Performance Summary
61
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
62
  |---|---|---|---|---|---|---
63
- | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® X Elite | 19.947 ms | 81 - 81 MB | NPU
64
- | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 14.9 ms | 0 - 483 MB | NPU
65
- | Conditional-DETR-ResNet50 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 19.711 ms | 0 - 95 MB | NPU
66
- | Conditional-DETR-ResNet50 | ONNX | float | Qualcomm® QCS9075 | 30.305 ms | 5 - 12 MB | NPU
67
- | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.03 ms | 4 - 397 MB | NPU
68
- | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.401 ms | 5 - 405 MB | NPU
69
- | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® X2 Elite | 8.902 ms | 82 - 82 MB | NPU
70
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 23.157 ms | 5 - 5 MB | NPU
71
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 16.79 ms | 0 - 428 MB | NPU
72
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 98.036 ms | 1 - 327 MB | NPU
73
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 22.728 ms | 5 - 8 MB | NPU
74
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 32.054 ms | 0 - 325 MB | NPU
75
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS9075 | 36.985 ms | 5 - 11 MB | NPU
76
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 46.556 ms | 3 - 372 MB | NPU
77
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 98.036 ms | 1 - 327 MB | NPU
78
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 34.228 ms | 0 - 281 MB | NPU
79
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.57 ms | 5 - 341 MB | NPU
80
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.777 ms | 5 - 347 MB | NPU
81
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 10.299 ms | 5 - 5 MB | NPU
82
- | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 16.962 ms | 0 - 463 MB | NPU
83
- | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 92.723 ms | 0 - 363 MB | NPU
84
- | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 22.327 ms | 0 - 3 MB | NPU
85
- | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® SA8775P | 138.15 ms | 0 - 421 MB | NPU
86
- | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS9075 | 33.268 ms | 0 - 93 MB | NPU
87
- | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 45.381 ms | 0 - 405 MB | NPU
88
- | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® SA7255P | 92.723 ms | 0 - 363 MB | NPU
89
- | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® SA8295P | 34.397 ms | 0 - 310 MB | NPU
90
- | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.883 ms | 0 - 368 MB | NPU
91
- | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 9.301 ms | 0 - 378 MB | NPU
92
 
93
  ## License
94
  * The license for the original implementation of Conditional-DETR-ResNet50 can be found
 
14
  DETR is a machine learning model that can detect objects (trained on COCO dataset).
15
 
16
  This is based on the implementation of Conditional-DETR-ResNet50 found [here](https://github.com/huggingface/transformers/tree/main/src/transformers/models/conditional_detr).
17
+ 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/main/qai_hub_models/models/conditional_detr_resnet50) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  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.
20
 
 
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
+ | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.48.0/conditional_detr_resnet50-onnx-float.zip)
31
+ | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.48.0/conditional_detr_resnet50-onnx-w8a16.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.48.0/conditional_detr_resnet50-qnn_dlc-float.zip)
33
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.48.0/conditional_detr_resnet50-qnn_dlc-w8a16.zip)
34
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.48.0/conditional_detr_resnet50-tflite-float.zip)
35
 
36
  For more device-specific assets and performance metrics, visit **[Conditional-DETR-ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/conditional_detr_resnet50)**.
37
 
38
 
39
  ### Option 2: Export with Custom Configurations
40
 
41
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/conditional_detr_resnet50) Python library to compile and export the model with your own:
42
  - Custom weights (e.g., fine-tuned checkpoints)
43
  - Custom input shapes
44
  - Target device and runtime configurations
45
 
46
  This option is ideal if you need to customize the model beyond the default configuration provided here.
47
 
48
+ See our repository for [Conditional-DETR-ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/conditional_detr_resnet50) for usage instructions.
49
 
50
  ## Model Details
51
 
 
60
  ## Performance Summary
61
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
62
  |---|---|---|---|---|---|---
63
+ | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® X2 Elite | 8.836 ms | 82 - 82 MB | NPU
64
+ | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® X Elite | 19.884 ms | 81 - 81 MB | NPU
65
+ | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 14.924 ms | 1 - 476 MB | NPU
66
+ | Conditional-DETR-ResNet50 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 19.763 ms | 0 - 96 MB | NPU
67
+ | Conditional-DETR-ResNet50 | ONNX | float | Qualcomm® QCS9075 | 30.848 ms | 5 - 12 MB | NPU
68
+ | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 10.98 ms | 2 - 397 MB | NPU
69
+ | Conditional-DETR-ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.364 ms | 5 - 406 MB | NPU
70
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 10.347 ms | 5 - 5 MB | NPU
71
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 23.183 ms | 5 - 5 MB | NPU
72
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 16.764 ms | 0 - 426 MB | NPU
73
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 97.911 ms | 1 - 324 MB | NPU
74
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 22.797 ms | 5 - 7 MB | NPU
75
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 32.053 ms | 0 - 325 MB | NPU
76
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS9075 | 34.054 ms | 5 - 11 MB | NPU
77
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 46.248 ms | 4 - 374 MB | NPU
78
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 97.911 ms | 1 - 324 MB | NPU
79
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 34.268 ms | 0 - 282 MB | NPU
80
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.564 ms | 5 - 341 MB | NPU
81
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.797 ms | 5 - 347 MB | NPU
82
+ | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 16.94 ms | 0 - 464 MB | NPU
83
+ | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 92.842 ms | 0 - 363 MB | NPU
84
+ | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 22.429 ms | 0 - 3 MB | NPU
85
+ | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® SA8775P | 30.187 ms | 0 - 423 MB | NPU
86
+ | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS9075 | 33.586 ms | 0 - 93 MB | NPU
87
+ | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 45.538 ms | 0 - 404 MB | NPU
88
+ | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® SA7255P | 92.842 ms | 0 - 363 MB | NPU
89
+ | Conditional-DETR-ResNet50 | TFLITE | float | Qualcomm® SA8295P | 34.36 ms | 0 - 311 MB | NPU
90
+ | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.909 ms | 0 - 375 MB | NPU
91
+ | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 9.315 ms | 0 - 380 MB | NPU
92
 
93
  ## License
94
  * The license for the original implementation of Conditional-DETR-ResNet50 can be found