Hugging Face's logo

qualcomm
/
BEVDet

Other
PyTorch
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
e87a31b
·
verified ·
1 Parent(s): 442cc4a

v0.59.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.59.0 for changelog.

Files changed (2) hide show
  1. README.md +132 -39
  2. release_assets.json +14 -7
README.md CHANGED
@@ -15,7 +15,7 @@ pipeline_tag: other
15
  BEVDet is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.
16
 
17
  This is based on the implementation of BEVDet found [here](https://github.com/HuangJunJie2017/BEVDet/).
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/bevdet) 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
 
@@ -28,22 +28,23 @@ Below are pre-exported model assets ready for deployment.
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
- | ONNX | float | Universal | ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.58.0/bevdet-onnx-float.zip)
32
- | TFLITE | float | Universal | | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.58.0/bevdet-tflite-float.zip)
 
33
 
34
  For more device-specific assets and performance metrics, visit **[BEVDet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/bevdet)**.
35
 
36
 
37
  ### Option 2: Export with Custom Configurations
38
 
39
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/bevdet) Python library to compile and export the model with your own:
40
  - Custom weights (e.g., fine-tuned checkpoints)
41
  - Custom input shapes
42
  - Target device and runtime configurations
43
 
44
  This option is ideal if you need to customize the model beyond the default configuration provided here.
45
 
46
- See our repository for [BEVDet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/bevdet) for usage instructions.
47
 
48
  ## Model Details
49
 
@@ -58,40 +59,132 @@ See our repository for [BEVDet on GitHub](https://github.com/qualcomm/ai-hub-mod
58
  ## Performance Summary
59
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
60
  |---|---|---|---|---|---|---
61
- | BEVDet | ONNX | float | Snapdragon® X2 Elite | 570.115 ms | 734 - 734 MB | CPU
62
- | BEVDet | ONNX | float | Snapdragon® X Elite | 3019.739 ms | 466 - 466 MB | CPU
63
- | BEVDet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 2358.617 ms | 211 - 221 MB | CPU
64
- | BEVDet | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 3077.073 ms | 204 - 219 MB | CPU
65
- | BEVDet | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2505.446 ms | 187 - 189 MB | CPU
66
- | BEVDet | ONNX | float | Qualcomm® QCS8450 | 3077.073 ms | 204 - 219 MB | CPU
67
- | BEVDet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1306.431 ms | 217 - 224 MB | CPU
68
- | BEVDet | ONNX | float | Snapdragon® 8 Elite Mobile | 1457.629 ms | 239 - 246 MB | CPU
69
- | BEVDet | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 1457.629 ms | 239 - 246 MB | CPU
70
- | BEVDet | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 3019.739 ms | 466 - 466 MB | CPU
71
- | BEVDet | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1506.065 ms | 241 - 255 MB | CPU
72
- | BEVDet | ONNX | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 724.814 ms | 712 - 712 MB | CPU
73
- | BEVDet | ONNX | w8a16_mixed_fp16 | Snapdragon® X Elite | 4253.027 ms | 709 - 709 MB | CPU
74
- | BEVDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 2437.909 ms | 303 - 317 MB | CPU
75
- | BEVDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 1 Mobile | 3113.118 ms | 418 - 440 MB | CPU
76
- | BEVDet | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2818.254 ms | 397 - 400 MB | CPU
77
- | BEVDet | ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8450 | 3113.118 ms | 418 - 440 MB | CPU
78
- | BEVDet | ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 1423.65 ms | 288 - 299 MB | CPU
79
- | BEVDet | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 1869.81 ms | 423 - 433 MB | CPU
80
- | BEVDet | ONNX | w8a16_mixed_fp16 | Qualcomm® Dragonwing™ IQ-X7181 | 4253.027 ms | 709 - 709 MB | CPU
81
- | BEVDet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1764.127 ms | 125 - 141 MB | CPU
82
- | BEVDet | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 2779.243 ms | 125 - 142 MB | CPU
83
- | BEVDet | TFLITE | float | Qualcomm® QCS8275 | 3163.413 ms | 128 - 137 MB | CPU
84
- | BEVDet | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2124.109 ms | 128 - 132 MB | CPU
85
- | BEVDet | TFLITE | float | Qualcomm® SA8775P | 2477.149 ms | 128 - 138 MB | CPU
86
- | BEVDet | TFLITE | float | Qualcomm® SA8650P | 2477.149 ms | 128 - 138 MB | CPU
87
- | BEVDet | TFLITE | float | Qualcomm® SA8255P | 2477.149 ms | 128 - 138 MB | CPU
88
- | BEVDet | TFLITE | float | Qualcomm® QCS8450 | 2779.243 ms | 125 - 142 MB | CPU
89
- | BEVDet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 389.525 ms | 127 - 140 MB | CPU
90
- | BEVDet | TFLITE | float | Snapdragon® 8 Elite Mobile | 1240.134 ms | 78 - 88 MB | CPU
91
- | BEVDet | TFLITE | float | Qualcomm® SA7255P | 3163.413 ms | 128 - 137 MB | CPU
92
- | BEVDet | TFLITE | float | Qualcomm® SA8295P | 1845.776 ms | 128 - 138 MB | CPU
93
- | BEVDet | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 1240.134 ms | 78 - 88 MB | CPU
94
- | BEVDet | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 2392.932 ms | 126 - 1330 MB | CPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
95
 
96
  ## License
97
  * The license for the original implementation of BEVDet can be found
 
15
  BEVDet is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.
16
 
17
  This is based on the implementation of BEVDet found [here](https://github.com/HuangJunJie2017/BEVDet/).
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/v0.59.0/src/qai_hub_models/models/bevdet) 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 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.59.0/bevdet-onnx-float.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.59.0/bevdet-qnn_dlc-float.zip)
33
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.59.0/bevdet-tflite-float.zip)
34
 
35
  For more device-specific assets and performance metrics, visit **[BEVDet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/bevdet)**.
36
 
37
 
38
  ### Option 2: Export with Custom Configurations
39
 
40
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/bevdet) Python library to compile and export the model with your own:
41
  - Custom weights (e.g., fine-tuned checkpoints)
42
  - Custom input shapes
43
  - Target device and runtime configurations
44
 
45
  This option is ideal if you need to customize the model beyond the default configuration provided here.
46
 
47
+ See our repository for [BEVDet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/bevdet) for usage instructions.
48
 
49
  ## Model Details
50
 
 
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
62
+ | decoder | ONNX | float | Snapdragon® X2 Elite | 6.414 ms | 4 - 4 MB | NPU
63
+ | decoder | ONNX | float | Snapdragon® X Elite | 12.433 ms | 34 - 34 MB | NPU
64
+ | decoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 8.829 ms | 0 - 173 MB | NPU
65
+ | decoder | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 23.478 ms | 0 - 170 MB | NPU
66
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.965 ms | 0 - 190 MB | NPU
67
+ | decoder | ONNX | float | Qualcomm® QCS8450 | 23.478 ms | 0 - 170 MB | NPU
68
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 23.694 ms | 4 - 10 MB | NPU
69
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 12.433 ms | 34 - 34 MB | NPU
70
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 7.63 ms | 3 - 150 MB | NPU
71
+ | decoder | ONNX | float | Snapdragon® 8 Elite Mobile | 7.63 ms | 3 - 150 MB | NPU
72
+ | decoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.569 ms | 3 - 155 MB | NPU
73
+ | decoder | QNN_DLC | float | Snapdragon® X2 Elite | 6.783 ms | 4 - 4 MB | NPU
74
+ | decoder | QNN_DLC | float | Snapdragon® X Elite | 12.525 ms | 4 - 4 MB | NPU
75
+ | decoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 8.808 ms | 0 - 170 MB | NPU
76
+ | decoder | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 23.259 ms | 0 - 170 MB | NPU
77
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 87.632 ms | 2 - 145 MB | NPU
78
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.884 ms | 4 - 6 MB | NPU
79
+ | decoder | QNN_DLC | float | Qualcomm® SA8775P | 23.126 ms | 1 - 149 MB | NPU
80
+ | decoder | QNN_DLC | float | Qualcomm® SA8650P | 23.126 ms | 1 - 149 MB | NPU
81
+ | decoder | QNN_DLC | float | Qualcomm® SA8255P | 23.126 ms | 1 - 149 MB | NPU
82
+ | decoder | QNN_DLC | float | Qualcomm® QCS8450 | 23.259 ms | 0 - 170 MB | NPU
83
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 23.436 ms | 6 - 13 MB | NPU
84
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 12.525 ms | 4 - 4 MB | NPU
85
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 7.585 ms | 0 - 146 MB | NPU
86
+ | decoder | QNN_DLC | float | Qualcomm® SA7255P | 87.632 ms | 2 - 145 MB | NPU
87
+ | decoder | QNN_DLC | float | Qualcomm® SA8295P | 25.559 ms | 2 - 144 MB | NPU
88
+ | decoder | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 7.585 ms | 0 - 146 MB | NPU
89
+ | decoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.57 ms | 4 - 154 MB | NPU
90
+ | decoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 8.814 ms | 1 - 203 MB | NPU
91
+ | decoder | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 23.352 ms | 1 - 206 MB | NPU
92
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 87.579 ms | 2 - 149 MB | NPU
93
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.951 ms | 1 - 3 MB | NPU
94
+ | decoder | TFLITE | float | Qualcomm® SA8775P | 23.142 ms | 1 - 153 MB | NPU
95
+ | decoder | TFLITE | float | Qualcomm® SA8650P | 23.142 ms | 1 - 153 MB | NPU
96
+ | decoder | TFLITE | float | Qualcomm® SA8255P | 23.142 ms | 1 - 153 MB | NPU
97
+ | decoder | TFLITE | float | Qualcomm® QCS8450 | 23.352 ms | 1 - 206 MB | NPU
98
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 23.632 ms | 0 - 44 MB | NPU
99
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 7.604 ms | 0 - 151 MB | NPU
100
+ | decoder | TFLITE | float | Qualcomm® SA7255P | 87.579 ms | 2 - 149 MB | NPU
101
+ | decoder | TFLITE | float | Qualcomm® SA8295P | 25.522 ms | 1 - 146 MB | NPU
102
+ | decoder | TFLITE | float | Snapdragon® 8 Elite Mobile | 7.604 ms | 0 - 151 MB | NPU
103
+ | decoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.547 ms | 1 - 152 MB | NPU
104
+ | encoder | ONNX | float | Snapdragon® X2 Elite | 21.086 ms | 12 - 12 MB | NPU
105
+ | encoder | ONNX | float | Snapdragon® X Elite | 39.992 ms | 40 - 40 MB | NPU
106
+ | encoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 30.749 ms | 2 - 419 MB | NPU
107
+ | encoder | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 70.927 ms | 1 - 328 MB | NPU
108
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 40.446 ms | 0 - 45 MB | NPU
109
+ | encoder | ONNX | float | Qualcomm® QCS8450 | 70.927 ms | 1 - 328 MB | NPU
110
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 66.703 ms | 12 - 28 MB | NPU
111
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 39.992 ms | 40 - 40 MB | NPU
112
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 25.191 ms | 5 - 326 MB | NPU
113
+ | encoder | ONNX | float | Snapdragon® 8 Elite Mobile | 25.191 ms | 5 - 326 MB | NPU
114
+ | encoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 18.329 ms | 4 - 334 MB | NPU
115
+ | encoder | QNN_DLC | float | Snapdragon® X2 Elite | 20.596 ms | 12 - 12 MB | NPU
116
+ | encoder | QNN_DLC | float | Snapdragon® X Elite | 40.278 ms | 12 - 12 MB | NPU
117
+ | encoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 30.737 ms | 12 - 421 MB | NPU
118
+ | encoder | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 70.707 ms | 2 - 326 MB | NPU
119
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 208.802 ms | 1 - 315 MB | NPU
120
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 40.977 ms | 12 - 577 MB | NPU
121
+ | encoder | QNN_DLC | float | Qualcomm® SA8775P | 62.539 ms | 1 - 311 MB | NPU
122
+ | encoder | QNN_DLC | float | Qualcomm® SA8650P | 62.539 ms | 1 - 311 MB | NPU
123
+ | encoder | QNN_DLC | float | Qualcomm® SA8255P | 62.539 ms | 1 - 311 MB | NPU
124
+ | encoder | QNN_DLC | float | Qualcomm® QCS8450 | 70.707 ms | 2 - 326 MB | NPU
125
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 62.965 ms | 12 - 28 MB | NPU
126
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 40.278 ms | 12 - 12 MB | NPU
127
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 25.277 ms | 0 - 315 MB | NPU
128
+ | encoder | QNN_DLC | float | Qualcomm® SA7255P | 208.802 ms | 1 - 315 MB | NPU
129
+ | encoder | QNN_DLC | float | Qualcomm® SA8295P | 62.603 ms | 0 - 226 MB | NPU
130
+ | encoder | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 25.277 ms | 0 - 315 MB | NPU
131
+ | encoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 18.285 ms | 12 - 338 MB | NPU
132
+ | encoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 33.359 ms | 1 - 454 MB | NPU
133
+ | encoder | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 70.639 ms | 2 - 366 MB | NPU
134
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 202.586 ms | 2 - 327 MB | NPU
135
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 38.942 ms | 2 - 5 MB | NPU
136
+ | encoder | TFLITE | float | Qualcomm® SA8775P | 59.601 ms | 2 - 327 MB | NPU
137
+ | encoder | TFLITE | float | Qualcomm® SA8650P | 59.601 ms | 2 - 327 MB | NPU
138
+ | encoder | TFLITE | float | Qualcomm® SA8255P | 59.601 ms | 2 - 327 MB | NPU
139
+ | encoder | TFLITE | float | Qualcomm® QCS8450 | 70.639 ms | 2 - 366 MB | NPU
140
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 60.404 ms | 0 - 66 MB | NPU
141
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 25.399 ms | 1 - 340 MB | NPU
142
+ | encoder | TFLITE | float | Qualcomm® SA7255P | 202.586 ms | 2 - 327 MB | NPU
143
+ | encoder | TFLITE | float | Qualcomm® SA8295P | 62.662 ms | 2 - 252 MB | NPU
144
+ | encoder | TFLITE | float | Snapdragon® 8 Elite Mobile | 25.399 ms | 1 - 340 MB | NPU
145
+ | encoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 19.309 ms | 2 - 351 MB | NPU
146
+ | pooler | ONNX | float | Snapdragon® X2 Elite | 21.694 ms | 535 - 535 MB | CPU
147
+ | pooler | ONNX | float | Snapdragon® X Elite | 111.062 ms | 534 - 534 MB | CPU
148
+ | pooler | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 28.212 ms | 334 - 342 MB | CPU
149
+ | pooler | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 40.059 ms | 324 - 339 MB | CPU
150
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 31.035 ms | 327 - 329 MB | CPU
151
+ | pooler | ONNX | float | Qualcomm® QCS8450 | 40.059 ms | 324 - 339 MB | CPU
152
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 56.799 ms | 330 - 334 MB | CPU
153
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 111.062 ms | 534 - 534 MB | CPU
154
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 28.814 ms | 327 - 336 MB | CPU
155
+ | pooler | ONNX | float | Snapdragon® 8 Elite Mobile | 28.814 ms | 327 - 336 MB | CPU
156
+ | pooler | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 22.744 ms | 332 - 340 MB | CPU
157
+ | pooler | QNN_DLC | float | Snapdragon® X2 Elite | 312.758 ms | 3 - 3 MB | CPU
158
+ | pooler | QNN_DLC | float | Snapdragon® X Elite | 343.215 ms | 3 - 3 MB | CPU
159
+ | pooler | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 155.796 ms | 3 - 14 MB | CPU
160
+ | pooler | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 201.461 ms | 2 - 19 MB | CPU
161
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 368.843 ms | 3 - 16 MB | CPU
162
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 175.362 ms | 4 - 7 MB | CPU
163
+ | pooler | QNN_DLC | float | Qualcomm® SA8775P | 320.679 ms | 4 - 16 MB | CPU
164
+ | pooler | QNN_DLC | float | Qualcomm® SA8650P | 320.679 ms | 4 - 16 MB | CPU
165
+ | pooler | QNN_DLC | float | Qualcomm® SA8255P | 320.679 ms | 4 - 16 MB | CPU
166
+ | pooler | QNN_DLC | float | Qualcomm® QCS8450 | 201.461 ms | 2 - 19 MB | CPU
167
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 332.407 ms | 270 - 741 MB | CPU
168
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 343.215 ms | 3 - 3 MB | CPU
169
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 177.66 ms | 4 - 13 MB | CPU
170
+ | pooler | QNN_DLC | float | Qualcomm® SA7255P | 368.843 ms | 3 - 16 MB | CPU
171
+ | pooler | QNN_DLC | float | Qualcomm® SA8295P | 308.142 ms | 4 - 16 MB | CPU
172
+ | pooler | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 177.66 ms | 4 - 13 MB | CPU
173
+ | pooler | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 214.683 ms | 4 - 14 MB | CPU
174
+ | pooler | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 55.944 ms | 126 - 139 MB | CPU
175
+ | pooler | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 90.138 ms | 127 - 142 MB | CPU
176
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 123.543 ms | 109 - 118 MB | CPU
177
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 74.295 ms | 128 - 130 MB | CPU
178
+ | pooler | TFLITE | float | Qualcomm® SA8775P | 106.547 ms | 129 - 134 MB | CPU
179
+ | pooler | TFLITE | float | Qualcomm® SA8650P | 106.547 ms | 129 - 134 MB | CPU
180
+ | pooler | TFLITE | float | Qualcomm® SA8255P | 106.547 ms | 129 - 134 MB | CPU
181
+ | pooler | TFLITE | float | Qualcomm® QCS8450 | 90.138 ms | 127 - 142 MB | CPU
182
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 107.311 ms | 127 - 320 MB | CPU
183
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 60.715 ms | 126 - 140 MB | CPU
184
+ | pooler | TFLITE | float | Qualcomm® SA7255P | 123.543 ms | 109 - 118 MB | CPU
185
+ | pooler | TFLITE | float | Qualcomm® SA8295P | 107.144 ms | 128 - 133 MB | CPU
186
+ | pooler | TFLITE | float | Snapdragon® 8 Elite Mobile | 60.715 ms | 126 - 140 MB | CPU
187
+ | pooler | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 53.035 ms | 127 - 139 MB | CPU
188
 
189
  ## License
190
  * The license for the original implementation of BEVDet can be found
release_assets.json CHANGED
@@ -1,19 +1,26 @@
1
  {
2
- "version": "0.58.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
- "tflite": {
7
  "tool_versions": {
8
- "litert": "1.4.4"
 
9
  },
10
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.58.0/bevdet-tflite-float.zip"
11
  },
12
- "onnx": {
 
 
 
 
 
 
13
  "tool_versions": {
14
- "onnx_runtime": "1.25.0"
15
  },
16
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.58.0/bevdet-onnx-float.zip"
17
  }
18
  }
19
  }
 
1
  {
2
+ "version": "0.59.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
+ "onnx": {
7
  "tool_versions": {
8
+ "qairt": "2.45.0.260326154327",
9
+ "onnx_runtime": "1.27.1"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.59.0/bevdet-onnx-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/bevdet/releases/v0.59.0/bevdet-qnn_dlc-float.zip"
18
+ },
19
+ "tflite": {
20
  "tool_versions": {
21
+ "qairt": "2.45.0.260326154327"
22
  },
23
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.59.0/bevdet-tflite-float.zip"
24
  }
25
  }
26
  }