Hugging Face's logo

qualcomm
/
Conditional-DETR-ResNet50

Object Detection
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
f4e4666
·
verified ·
1 Parent(s): fb6ae94

v0.64.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.64.0 for changelog.

Files changed (2) hide show
  1. README.md +34 -37
  2. release_assets.json +6 -6
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/qualcomm/ai-hub-models/blob/v0.63.0/src/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 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-onnx-float.zip)
31
- | ONNX | w8a16_mixed_fp16 | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-onnx-w8a16_mixed_fp16.zip)
32
- | QNN_DLC | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-qnn_dlc-float.zip)
33
- | QNN_DLC | w8a16_mixed_fp16 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-qnn_dlc-w8a16_mixed_fp16.zip)
34
- | TFLITE | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/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/v0.63.0/src/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/v0.63.0/src/qai_hub_models/models/conditional_detr_resnet50) for usage instructions.
49
 
50
  ## Model Details
51
 
@@ -76,35 +76,32 @@ See our repository for [Conditional-DETR-ResNet50 on GitHub](https://github.com/
76
  | Conditional-DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 41.47 ms | 24 - 287 MB | NPU
77
  | Conditional-DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 47.381 ms | 21 - 394 MB | NPU
78
  | Conditional-DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 41.47 ms | 24 - 287 MB | NPU
79
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 9.283 ms | 5 - 320 MB | NPU
80
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.406 ms | 0 - 301 MB | NPU
81
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 10.809 ms | 5 - 5 MB | NPU
82
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 22.926 ms | 5 - 5 MB | NPU
83
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 16.667 ms | 5 - 389 MB | NPU
84
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 32.234 ms | 5 - 340 MB | NPU
85
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 33.194 ms | 5 - 12 MB | NPU
86
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 22.628 ms | 5 - 7 MB | NPU
87
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 31.601 ms | 0 - 292 MB | NPU
88
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8650P | 31.601 ms | 0 - 292 MB | NPU
89
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8255P | 31.601 ms | 0 - 292 MB | NPU
90
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 32.234 ms | 5 - 340 MB | NPU
91
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 32.365 ms | 5 - 12 MB | NPU
92
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 22.926 ms | 5 - 5 MB | NPU
93
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 11.406 ms | 0 - 301 MB | NPU
94
- | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 34.446 ms | 0 - 258 MB | NPU
95
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 10.95 ms | 2 - 343 MB | NPU
96
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 13.569 ms | 2 - 331 MB | NPU
97
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 10.832 ms | 2 - 2 MB | NPU
98
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X Elite | 26.069 ms | 2 - 2 MB | NPU
99
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 18.894 ms | 2 - 418 MB | NPU
100
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-8275 | 31.253 ms | 2 - 8 MB | NPU
101
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 25.991 ms | 2 - 4 MB | NPU
102
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8650P | 35.951 ms | 1 - 370 MB | NPU
103
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8255P | 35.951 ms | 1 - 370 MB | NPU
104
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 31.163 ms | 2 - 7 MB | NPU
105
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-X7181 | 26.069 ms | 2 - 2 MB | NPU
106
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 13.569 ms | 2 - 331 MB | NPU
107
- | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA7255P | 80.624 ms | 0 - 322 MB | NPU
108
  | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 9.578 ms | 0 - 344 MB | NPU
109
  | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 12.277 ms | 0 - 333 MB | NPU
110
  | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 16.117 ms | 0 - 406 MB | NPU
 
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/v0.64.0/src/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 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-onnx-float.zip)
31
+ | ONNX | w8a16_mixed_fp16 | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-onnx-w8a16_mixed_fp16.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-qnn_dlc-float.zip)
33
+ | QNN_DLC | w8a16_mixed_fp16 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-qnn_dlc-w8a16_mixed_fp16.zip)
34
+ | TFLITE | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/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/v0.64.0/src/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/v0.64.0/src/qai_hub_models/models/conditional_detr_resnet50) for usage instructions.
49
 
50
  ## Model Details
51
 
 
76
  | Conditional-DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 41.47 ms | 24 - 287 MB | NPU
77
  | Conditional-DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 47.381 ms | 21 - 394 MB | NPU
78
  | Conditional-DETR-ResNet50 | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 41.47 ms | 24 - 287 MB | NPU
79
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 9.283 ms | 4 - 319 MB | NPU
80
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.405 ms | 5 - 305 MB | NPU
81
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 10.758 ms | 5 - 5 MB | NPU
82
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 22.941 ms | 5 - 5 MB | NPU
83
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 16.667 ms | 3 - 387 MB | NPU
84
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 31.9 ms | 5 - 340 MB | NPU
85
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 33.278 ms | 5 - 12 MB | NPU
86
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 22.537 ms | 5 - 7 MB | NPU
87
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 31.9 ms | 5 - 340 MB | NPU
88
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 32.093 ms | 7 - 14 MB | NPU
89
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 22.941 ms | 5 - 5 MB | NPU
90
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 11.405 ms | 5 - 305 MB | NPU
91
+ | Conditional-DETR-ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 34.473 ms | 0 - 258 MB | NPU
92
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 10.962 ms | 2 - 342 MB | NPU
93
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 13.601 ms | 2 - 331 MB | NPU
94
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 10.669 ms | 2 - 2 MB | NPU
95
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® X Elite | 26.106 ms | 2 - 2 MB | NPU
96
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 19.02 ms | 2 - 419 MB | NPU
97
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-8275 | 31.109 ms | 2 - 8 MB | NPU
98
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 25.829 ms | 2 - 4 MB | NPU
99
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8775P | 30.871 ms | 1 - 324 MB | NPU
100
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8650P | 30.871 ms | 1 - 324 MB | NPU
101
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® SA8255P | 30.871 ms | 1 - 324 MB | NPU
102
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 31.134 ms | 2 - 7 MB | NPU
103
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-X7181 | 26.106 ms | 2 - 2 MB | NPU
104
+ | Conditional-DETR-ResNet50 | QNN_DLC | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 13.601 ms | 2 - 331 MB | NPU
 
 
 
105
  | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 9.578 ms | 0 - 344 MB | NPU
106
  | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 12.277 ms | 0 - 333 MB | NPU
107
  | Conditional-DETR-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 16.117 ms | 0 - 406 MB | NPU
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.63.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
@@ -8,19 +8,19 @@
8
  "qairt": "2.50.0.260828221209",
9
  "onnx_runtime": "1.27.1"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.50.0.260828221209"
16
  },
17
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.50.0.260828221209"
22
  },
23
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-tflite-float.zip"
24
  }
25
  }
26
  },
@@ -31,13 +31,13 @@
31
  "qairt": "2.50.0.260828221209",
32
  "onnx_runtime": "1.27.1"
33
  },
34
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-onnx-w8a16_mixed_fp16.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.50.0.260828221209"
39
  },
40
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.63.0/conditional_detr_resnet50-qnn_dlc-w8a16_mixed_fp16.zip"
41
  }
42
  }
43
  }
 
1
  {
2
+ "version": "0.64.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
 
8
  "qairt": "2.50.0.260828221209",
9
  "onnx_runtime": "1.27.1"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.50.0.260828221209"
16
  },
17
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.50.0.260828221209"
22
  },
23
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-tflite-float.zip"
24
  }
25
  }
26
  },
 
31
  "qairt": "2.50.0.260828221209",
32
  "onnx_runtime": "1.27.1"
33
  },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-onnx-w8a16_mixed_fp16.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.50.0.260828221209"
39
  },
40
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/conditional_detr_resnet50/releases/v0.64.0/conditional_detr_resnet50-qnn_dlc-w8a16_mixed_fp16.zip"
41
  }
42
  }
43
  }