Hugging Face's logo

qualcomm
/
Yolo-R

Object Detection
PyTorch
bu_auto
real_time
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
8b32ba9
·
verified ·
1 Parent(s): 12759fe

v0.58.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.58.0 for changelog.

Files changed (1) hide show
  1. README.md +46 -43
README.md CHANGED
@@ -16,18 +16,18 @@ pipeline_tag: object-detection
16
  YoloR is a machine learning model that predicts bounding boxes and classes of objects in an image.
17
 
18
  This is based on the implementation of Yolo-R found [here](https://github.com/WongKinYiu/yolor).
19
- 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/v0.57.3/src/qai_hub_models/models/yolor) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
20
 
21
  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.
22
 
23
  ## Getting Started
24
  Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
25
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.57.3/src/qai_hub_models/models/yolor) Python library to compile and export the model with your own:
26
  - Custom weights (e.g., fine-tuned checkpoints)
27
  - Custom input shapes
28
  - Target device and runtime configurations
29
 
30
- See our repository for [Yolo-R on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.57.3/src/qai_hub_models/models/yolor) for usage instructions.
31
 
32
 
33
  ## Model Details
@@ -43,46 +43,49 @@ See our repository for [Yolo-R on GitHub](https://github.com/qualcomm/ai-hub-mod
43
  ## Performance Summary
44
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
45
  |---|---|---|---|---|---|---
46
- | Yolo-R | ONNX | float | Snapdragon® X2 Elite | 26.281 ms | 208 - 208 MB | NPU
47
- | Yolo-R | ONNX | float | Snapdragon® X Elite | 48.569 ms | 144 - 144 MB | NPU
48
- | Yolo-R | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 37.375 ms | 6 - 325 MB | NPU
49
- | Yolo-R | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 62.973 ms | 6 - 374 MB | NPU
50
- | Yolo-R | ONNX | float | Qualcomm® QCS8550 (Proxy) | 48.26 ms | 0 - 80 MB | NPU
51
- | Yolo-R | ONNX | float | Qualcomm® QCS8450 | 62.973 ms | 6 - 374 MB | NPU
52
- | Yolo-R | ONNX | float | Snapdragon® 8 Elite Mobile | 27.253 ms | 1 - 231 MB | NPU
53
- | Yolo-R | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 26.587 ms | 2 - 292 MB | NPU
54
- | Yolo-R | ONNX | float | Qualcomm® QCS9075 | 54.75 ms | 5 - 50 MB | NPU
55
- | Yolo-R | ONNX | float | Qualcomm® QCS8750 | 27.253 ms | 1 - 231 MB | NPU
56
- | Yolo-R | ONNX | float | Qualcomm® QCS7181 | 48.569 ms | 144 - 144 MB | NPU
57
- | Yolo-R | ONNX | w8a16 | Snapdragon® X2 Elite | 18.191 ms | 179 - 179 MB | NPU
58
- | Yolo-R | ONNX | w8a16 | Snapdragon® X Elite | 30.065 ms | 148 - 148 MB | NPU
59
- | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 20.838 ms | 0 - 448 MB | NPU
60
- | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 38.253 ms | 4 - 452 MB | NPU
61
- | Yolo-R | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 28.747 ms | 0 - 53 MB | NPU
62
- | Yolo-R | ONNX | w8a16 | Qualcomm® QCS8450 | 38.253 ms | 4 - 452 MB | NPU
63
- | Yolo-R | ONNX | w8a16 | Qualcomm® QCS9075 | 29.573 ms | 2 - 48 MB | NPU
64
- | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 17.801 ms | 1 - 421 MB | NPU
65
- | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 17.071 ms | 1 - 365 MB | NPU
66
- | Yolo-R | ONNX | w8a16 | Qualcomm® QCS8750 | 17.071 ms | 1 - 365 MB | NPU
67
- | Yolo-R | ONNX | w8a16 | Qualcomm® QCS7181 | 30.065 ms | 148 - 148 MB | NPU
68
- | Yolo-R | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 8.757 ms | 2 - 2 MB | NPU
69
- | Yolo-R | QNN_DLC | w8a16 | Snapdragon® X Elite | 20.361 ms | 2 - 2 MB | NPU
70
- | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 13.118 ms | 2 - 358 MB | NPU
71
- | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Gen 1 Mobile | 27.847 ms | 2 - 359 MB | NPU
72
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 74.958 ms | 2 - 7 MB | NPU
73
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS8275 | 39.545 ms | 1 - 293 MB | NPU
74
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 19.601 ms | 2 - 5 MB | NPU
75
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS8450 | 27.847 ms | 2 - 359 MB | NPU
76
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 20.366 ms | 2 - 7 MB | NPU
77
- | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 7.717 ms | 2 - 311 MB | NPU
78
- | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 9.815 ms | 2 - 305 MB | NPU
79
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® SA8295P | 25.044 ms | 0 - 294 MB | NPU
80
- | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 25.938 ms | 2 - 312 MB | NPU
81
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® SA7255P | 39.545 ms | 1 - 293 MB | NPU
82
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 222.028 ms | 2 - 398 MB | NPU
83
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS7790 | 25.938 ms | 2 - 312 MB | NPU
84
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS8750 | 9.815 ms | 2 - 305 MB | NPU
85
- | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS7181 | 20.361 ms | 2 - 2 MB | NPU
 
 
 
86
 
87
  ## License
88
  * The license for the original implementation of Yolo-R can be found
 
16
  YoloR is a machine learning model that predicts bounding boxes and classes of objects in an image.
17
 
18
  This is based on the implementation of Yolo-R found [here](https://github.com/WongKinYiu/yolor).
19
+ 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/yolor) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
20
 
21
  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.
22
 
23
  ## Getting Started
24
  Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
25
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/yolor) Python library to compile and export the model with your own:
26
  - Custom weights (e.g., fine-tuned checkpoints)
27
  - Custom input shapes
28
  - Target device and runtime configurations
29
 
30
+ See our repository for [Yolo-R on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/yolor) for usage instructions.
31
 
32
 
33
  ## Model Details
 
43
  ## Performance Summary
44
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
45
  |---|---|---|---|---|---|---
46
+ | Yolo-R | ONNX | float | Snapdragon® X2 Elite | 26.434 ms | 5 - 5 MB | NPU
47
+ | Yolo-R | ONNX | float | Snapdragon® X Elite | 48.185 ms | 75 - 75 MB | NPU
48
+ | Yolo-R | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 37.545 ms | 4 - 329 MB | NPU
49
+ | Yolo-R | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 62.726 ms | 3 - 380 MB | NPU
50
+ | Yolo-R | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 48.235 ms | 0 - 519 MB | NPU
51
+ | Yolo-R | ONNX | float | Qualcomm® QCS8450 | 62.726 ms | 3 - 380 MB | NPU
52
+ | Yolo-R | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 70.779 ms | 5 - 12 MB | NPU
53
+ | Yolo-R | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 26.781 ms | 2 - 296 MB | NPU
54
+ | Yolo-R | ONNX | float | Snapdragon® 8 Elite Mobile | 27.096 ms | 3 - 234 MB | NPU
55
+ | Yolo-R | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 27.096 ms | 3 - 234 MB | NPU
56
+ | Yolo-R | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 48.185 ms | 75 - 75 MB | NPU
57
+ | Yolo-R | ONNX | w8a16 | Snapdragon® X2 Elite | 18.288 ms | 2 - 2 MB | NPU
58
+ | Yolo-R | ONNX | w8a16 | Snapdragon® X Elite | 30.005 ms | 41 - 41 MB | NPU
59
+ | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 20.87 ms | 3 - 461 MB | NPU
60
+ | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 38.245 ms | 3 - 462 MB | NPU
61
+ | Yolo-R | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 29.08 ms | 0 - 49 MB | NPU
62
+ | Yolo-R | ONNX | w8a16 | Qualcomm® QCS8450 | 38.245 ms | 3 - 462 MB | NPU
63
+ | Yolo-R | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 30.142 ms | 1 - 6 MB | NPU
64
+ | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 17.939 ms | 1 - 424 MB | NPU
65
+ | Yolo-R | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 17.178 ms | 1 - 363 MB | NPU
66
+ | Yolo-R | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 17.178 ms | 1 - 363 MB | NPU
67
+ | Yolo-R | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 30.005 ms | 41 - 41 MB | NPU
68
+ | Yolo-R | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 8.733 ms | 2 - 2 MB | NPU
69
+ | Yolo-R | QNN_DLC | w8a16 | Snapdragon® X Elite | 20.385 ms | 2 - 2 MB | NPU
70
+ | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 13.158 ms | 2 - 356 MB | NPU
71
+ | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Gen 1 Mobile | 28.084 ms | 0 - 356 MB | NPU
72
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 90.79 ms | 2 - 7 MB | NPU
73
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS8275 | 39.521 ms | 2 - 293 MB | NPU
74
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 19.436 ms | 2 - 5 MB | NPU
75
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® SA8775P | 19.635 ms | 1 - 291 MB | NPU
76
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® SA8650P | 19.635 ms | 1 - 291 MB | NPU
77
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® SA8255P | 19.635 ms | 1 - 291 MB | NPU
78
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® QCS8450 | 28.084 ms | 0 - 356 MB | NPU
79
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 19.998 ms | 2 - 6 MB | NPU
80
+ | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 7.758 ms | 2 - 311 MB | NPU
81
+ | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 25.922 ms | 2 - 315 MB | NPU
82
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® SA7255P | 39.521 ms | 2 - 293 MB | NPU
83
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 222.304 ms | 2 - 397 MB | NPU
84
+ | Yolo-R | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 9.855 ms | 2 - 304 MB | NPU
85
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® SA8295P | 25.073 ms | 0 - 294 MB | NPU
86
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 25.922 ms | 2 - 315 MB | NPU
87
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 9.855 ms | 2 - 304 MB | NPU
88
+ | Yolo-R | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 20.385 ms | 2 - 2 MB | NPU
89
 
90
  ## License
91
  * The license for the original implementation of Yolo-R can be found