Hugging Face's logo

qualcomm
/
ResNet50

Image Classification
PyTorch
backbone
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
a9488bd
·
verified ·
1 Parent(s): 4a62740

v0.58.0

Browse files

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

Files changed (2) hide show
  1. README.md +103 -97
  2. release_assets.json +7 -7
README.md CHANGED
@@ -16,7 +16,7 @@ pipeline_tag: image-classification
16
  ResNet50 is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.
17
 
18
  This is based on the implementation of ResNet50 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py).
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/resnet50) 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
 
@@ -29,26 +29,26 @@ Below are pre-exported model assets ready for deployment.
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
- | 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/resnet50/releases/v0.57.3/resnet50-onnx-float.zip)
33
- | ONNX | w8a8 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-onnx-w8a8.zip)
34
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-qnn_dlc-float.zip)
35
- | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-qnn_dlc-w8a8.zip)
36
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-tflite-float.zip)
37
- | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-tflite-w8a8.zip)
38
 
39
  For more device-specific assets and performance metrics, visit **[ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/resnet50)**.
40
 
41
 
42
  ### Option 2: Export with Custom Configurations
43
 
44
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.57.3/src/qai_hub_models/models/resnet50) Python library to compile and export the model with your own:
45
  - Custom weights (e.g., fine-tuned checkpoints)
46
  - Custom input shapes
47
  - Target device and runtime configurations
48
 
49
  This option is ideal if you need to customize the model beyond the default configuration provided here.
50
 
51
- See our repository for [ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.57.3/src/qai_hub_models/models/resnet50) for usage instructions.
52
 
53
  ## Model Details
54
 
@@ -64,96 +64,102 @@ See our repository for [ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-m
64
  ## Performance Summary
65
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
66
  |---|---|---|---|---|---|---
67
- | ResNet50 | ONNX | float | Snapdragon® X2 Elite | 0.982 ms | 180 - 180 MB | NPU
68
- | ResNet50 | ONNX | float | Snapdragon® X Elite | 1.914 ms | 181 - 181 MB | NPU
69
- | ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.471 ms | 1 - 71 MB | NPU
70
- | ResNet50 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 3.18 ms | 1 - 54 MB | NPU
71
- | ResNet50 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 1.979 ms | 0 - 60 MB | NPU
72
- | ResNet50 | ONNX | float | Qualcomm® QCS8450 | 3.18 ms | 1 - 54 MB | NPU
73
- | ResNet50 | ONNX | float | Snapdragon® 8 Elite Mobile | 1.218 ms | 0 - 54 MB | NPU
74
- | ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.057 ms | 1 - 49 MB | NPU
75
- | ResNet50 | ONNX | float | Qualcomm® QCS9075 | 3.168 ms | 1 - 46 MB | NPU
76
- | ResNet50 | ONNX | float | Qualcomm® QCS8750 | 1.218 ms | 0 - 54 MB | NPU
77
- | ResNet50 | ONNX | float | Qualcomm® QCS7181 | 1.914 ms | 181 - 181 MB | NPU
78
- | ResNet50 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.415 ms | 181 - 181 MB | NPU
79
- | ResNet50 | ONNX | w8a8 | Snapdragon® X Elite | 0.831 ms | 181 - 181 MB | NPU
80
- | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.687 ms | 0 - 75 MB | NPU
81
- | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.216 ms | 0 - 75 MB | NPU
82
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCS6490 | 2.884 ms | 0 - 45 MB | NPU
83
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.91 ms | 0 - 35 MB | NPU
84
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCS8450 | 1.216 ms | 0 - 75 MB | NPU
85
- | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.558 ms | 0 - 46 MB | NPU
86
- | ResNet50 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.202 ms | 0 - 51 MB | NPU
87
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCM6690 | 6.481 ms | 0 - 169 MB | NPU
88
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCS9075 | 0.991 ms | 0 - 45 MB | NPU
89
- | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.587 ms | 0 - 50 MB | NPU
90
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCS7790 | 1.202 ms | 0 - 51 MB | NPU
91
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCS8750 | 0.587 ms | 0 - 50 MB | NPU
92
- | ResNet50 | ONNX | w8a8 | Qualcomm® QCS7181 | 0.831 ms | 181 - 181 MB | NPU
93
- | ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 1.219 ms | 1 - 1 MB | NPU
94
- | ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 2.327 ms | 1 - 1 MB | NPU
95
- | ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.621 ms | 0 - 76 MB | NPU
96
- | ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 3.727 ms | 0 - 61 MB | NPU
97
- | ResNet50 | QNN_DLC | float | Qualcomm® QCS8275 | 10.725 ms | 1 - 46 MB | NPU
98
- | ResNet50 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 2.242 ms | 1 - 136 MB | NPU
99
- | ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 3.727 ms | 0 - 61 MB | NPU
100
- | ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 1.262 ms | 1 - 51 MB | NPU
101
- | ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 3.668 ms | 0 - 28 MB | NPU
102
- | ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.09 ms | 1 - 48 MB | NPU
103
- | ResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 10.725 ms | 1 - 46 MB | NPU
104
- | ResNet50 | QNN_DLC | float | Qualcomm® QCS9075 | 3.298 ms | 3 - 5 MB | NPU
105
- | ResNet50 | QNN_DLC | float | Qualcomm® QCS8750 | 1.262 ms | 1 - 51 MB | NPU
106
- | ResNet50 | QNN_DLC | float | Qualcomm® QCS7181 | 2.327 ms | 1 - 1 MB | NPU
107
- | ResNet50 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.468 ms | 0 - 0 MB | NPU
108
- | ResNet50 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.933 ms | 0 - 0 MB | NPU
109
- | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.683 ms | 0 - 77 MB | NPU
110
- | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.193 ms | 0 - 78 MB | NPU
111
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 2.986 ms | 0 - 2 MB | NPU
112
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8275 | 2.015 ms | 0 - 45 MB | NPU
113
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.906 ms | 0 - 6 MB | NPU
114
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 1.193 ms | 0 - 78 MB | NPU
115
- | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.488 ms | 0 - 44 MB | NPU
116
- | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.223 ms | 0 - 51 MB | NPU
117
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 6.616 ms | 0 - 164 MB | NPU
118
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.968 ms | 0 - 2 MB | NPU
119
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 2.015 ms | 0 - 45 MB | NPU
120
- | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.525 ms | 0 - 42 MB | NPU
121
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.442 ms | 0 - 42 MB | NPU
122
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS7790 | 1.223 ms | 0 - 51 MB | NPU
123
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8750 | 0.525 ms | 0 - 42 MB | NPU
124
- | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS7181 | 0.933 ms | 0 - 0 MB | NPU
125
- | ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.624 ms | 0 - 115 MB | NPU
126
- | ResNet50 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 3.745 ms | 0 - 95 MB | NPU
 
 
 
 
 
 
127
  | ResNet50 | TFLITE | float | Qualcomm® QCS8275 | 10.663 ms | 0 - 69 MB | NPU
128
- | ResNet50 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 2.206 ms | 0 - 4 MB | NPU
129
- | ResNet50 | TFLITE | float | Qualcomm® SA8775P | 20.779 ms | 0 - 21 MB | GPU
130
- | ResNet50 | TFLITE | float | Qualcomm® SA8650P | 20.779 ms | 0 - 21 MB | GPU
131
- | ResNet50 | TFLITE | float | Qualcomm® SA8255P | 20.779 ms | 0 - 21 MB | GPU
132
- | ResNet50 | TFLITE | float | Qualcomm® QCS8450 | 3.745 ms | 0 - 95 MB | NPU
133
- | ResNet50 | TFLITE | float | Snapdragon® 8 Elite Mobile | 1.249 ms | 0 - 67 MB | NPU
134
- | ResNet50 | TFLITE | float | Qualcomm® SA8295P | 3.638 ms | 0 - 50 MB | NPU
135
- | ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.092 ms | 0 - 66 MB | NPU
136
  | ResNet50 | TFLITE | float | Qualcomm® SA7255P | 10.663 ms | 0 - 69 MB | NPU
137
- | ResNet50 | TFLITE | float | Qualcomm® QCS9075 | 3.338 ms | 0 - 52 MB | NPU
138
- | ResNet50 | TFLITE | float | Qualcomm® QCS8750 | 1.249 ms | 0 - 67 MB | NPU
139
- | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.576 ms | 0 - 74 MB | NPU
140
- | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.05 ms | 0 - 80 MB | NPU
141
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS6490 | 2.629 ms | 0 - 27 MB | NPU
142
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8275 | 1.775 ms | 0 - 42 MB | NPU
143
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.767 ms | 0 - 3 MB | NPU
144
- | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8775P | 21.203 ms | 1 - 25 MB | GPU
145
- | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8650P | 21.203 ms | 1 - 25 MB | GPU
146
- | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8255P | 21.203 ms | 1 - 25 MB | GPU
147
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8450 | 1.05 ms | 0 - 80 MB | NPU
148
- | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.448 ms | 0 - 42 MB | NPU
149
- | ResNet50 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.095 ms | 0 - 49 MB | NPU
150
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCM6690 | 6.164 ms | 0 - 161 MB | NPU
151
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.821 ms | 0 - 27 MB | NPU
152
- | ResNet50 | TFLITE | w8a8 | Qualcomm® SA7255P | 1.775 ms | 0 - 42 MB | NPU
153
- | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.495 ms | 0 - 44 MB | NPU
154
- | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8295P | 1.295 ms | 0 - 39 MB | NPU
155
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS7790 | 1.095 ms | 0 - 49 MB | NPU
156
- | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8750 | 0.495 ms | 0 - 44 MB | NPU
 
157
 
158
  ## License
159
  * The license for the original implementation of ResNet50 can be found
 
16
  ResNet50 is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.
17
 
18
  This is based on the implementation of ResNet50 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py).
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/resnet50) 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
 
 
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
+ | 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/resnet50/releases/v0.58.0/resnet50-onnx-float.zip)
33
+ | ONNX | w8a8 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-onnx-w8a8.zip)
34
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-qnn_dlc-float.zip)
35
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-qnn_dlc-w8a8.zip)
36
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-tflite-float.zip)
37
+ | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-tflite-w8a8.zip)
38
 
39
  For more device-specific assets and performance metrics, visit **[ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/resnet50)**.
40
 
41
 
42
  ### Option 2: Export with Custom Configurations
43
 
44
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/resnet50) Python library to compile and export the model with your own:
45
  - Custom weights (e.g., fine-tuned checkpoints)
46
  - Custom input shapes
47
  - Target device and runtime configurations
48
 
49
  This option is ideal if you need to customize the model beyond the default configuration provided here.
50
 
51
+ See our repository for [ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/resnet50) for usage instructions.
52
 
53
  ## Model Details
54
 
 
64
  ## Performance Summary
65
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
66
  |---|---|---|---|---|---|---
67
+ | ResNet50 | ONNX | float | Snapdragon® X2 Elite | 0.983 ms | 2 - 2 MB | NPU
68
+ | ResNet50 | ONNX | float | Snapdragon® X Elite | 1.925 ms | 50 - 50 MB | NPU
69
+ | ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.464 ms | 0 - 80 MB | NPU
70
+ | ResNet50 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 3.194 ms | 0 - 62 MB | NPU
71
+ | ResNet50 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.985 ms | 0 - 70 MB | NPU
72
+ | ResNet50 | ONNX | float | Qualcomm® QCS8450 | 3.194 ms | 0 - 62 MB | NPU
73
+ | ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 3.018 ms | 0 - 4 MB | NPU
74
+ | ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.058 ms | 0 - 53 MB | NPU
75
+ | ResNet50 | ONNX | float | Snapdragon® 8 Elite Mobile | 1.207 ms | 0 - 53 MB | NPU
76
+ | ResNet50 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 1.207 ms | 0 - 53 MB | NPU
77
+ | ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.925 ms | 50 - 50 MB | NPU
78
+ | ResNet50 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.41 ms | 1 - 1 MB | NPU
79
+ | ResNet50 | ONNX | w8a8 | Snapdragon® X Elite | 0.828 ms | 25 - 25 MB | NPU
80
+ | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.681 ms | 0 - 85 MB | NPU
81
+ | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.21 ms | 0 - 84 MB | NPU
82
+ | ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 3.725 ms | 0 - 3 MB | NPU
83
+ | ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.906 ms | 0 - 67 MB | NPU
84
+ | ResNet50 | ONNX | w8a8 | Qualcomm® QCS8450 | 1.21 ms | 0 - 84 MB | NPU
85
+ | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.581 ms | 0 - 45 MB | NPU
86
+ | ResNet50 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.202 ms | 0 - 55 MB | NPU
87
+ | ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.55 ms | 0 - 50 MB | NPU
88
+ | ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.005 ms | 0 - 3 MB | NPU
89
+ | ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 6.469 ms | 0 - 168 MB | NPU
90
+ | ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.202 ms | 0 - 55 MB | NPU
91
+ | ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.581 ms | 0 - 45 MB | NPU
92
+ | ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.828 ms | 25 - 25 MB | NPU
93
+ | ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 1.234 ms | 1 - 1 MB | NPU
94
+ | ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 2.314 ms | 1 - 1 MB | NPU
95
+ | ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.607 ms | 1 - 79 MB | NPU
96
+ | ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 3.742 ms | 0 - 63 MB | NPU
97
+ | ResNet50 | QNN_DLC | float | Qualcomm® QCS8275 | 10.72 ms | 1 - 46 MB | NPU
98
+ | ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2.202 ms | 0 - 3 MB | NPU
99
+ | ResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 3.359 ms | 1 - 47 MB | NPU
100
+ | ResNet50 | QNN_DLC | float | Qualcomm® SA8650P | 3.359 ms | 1 - 47 MB | NPU
101
+ | ResNet50 | QNN_DLC | float | Qualcomm® SA8255P | 3.359 ms | 1 - 47 MB | NPU
102
+ | ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 3.742 ms | 0 - 63 MB | NPU
103
+ | ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 3.18 ms | 3 - 5 MB | NPU
104
+ | ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.096 ms | 1 - 48 MB | NPU
105
+ | ResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 10.72 ms | 1 - 46 MB | NPU
106
+ | ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 1.263 ms | 0 - 48 MB | NPU
107
+ | ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 3.659 ms | 0 - 28 MB | NPU
108
+ | ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 1.263 ms | 0 - 48 MB | NPU
109
+ | ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 2.314 ms | 1 - 1 MB | NPU
110
+ | ResNet50 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.45 ms | 0 - 0 MB | NPU
111
+ | ResNet50 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.937 ms | 0 - 0 MB | NPU
112
+ | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.683 ms | 0 - 74 MB | NPU
113
+ | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.186 ms | 0 - 76 MB | NPU
114
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 3.768 ms | 2 - 4 MB | NPU
115
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8275 | 2.012 ms | 0 - 45 MB | NPU
116
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.898 ms | 0 - 2 MB | NPU
117
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 1.084 ms | 0 - 47 MB | NPU
118
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8650P | 1.084 ms | 0 - 47 MB | NPU
119
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8255P | 1.084 ms | 0 - 47 MB | NPU
120
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 1.186 ms | 0 - 76 MB | NPU
121
+ | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.534 ms | 0 - 42 MB | NPU
122
+ | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.221 ms | 0 - 51 MB | NPU
123
+ | ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.484 ms | 0 - 48 MB | NPU
124
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.985 ms | 0 - 2 MB | NPU
125
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 6.619 ms | 0 - 163 MB | NPU
126
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 2.012 ms | 0 - 45 MB | NPU
127
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.459 ms | 0 - 42 MB | NPU
128
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.221 ms | 0 - 51 MB | NPU
129
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.534 ms | 0 - 42 MB | NPU
130
+ | ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.937 ms | 0 - 0 MB | NPU
131
+ | ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.585 ms | 0 - 111 MB | NPU
132
+ | ResNet50 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 3.727 ms | 0 - 96 MB | NPU
133
  | ResNet50 | TFLITE | float | Qualcomm® QCS8275 | 10.663 ms | 0 - 69 MB | NPU
134
+ | ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2.206 ms | 0 - 3 MB | NPU
135
+ | ResNet50 | TFLITE | float | Qualcomm® SA8775P | 3.343 ms | 0 - 70 MB | NPU
136
+ | ResNet50 | TFLITE | float | Qualcomm® SA8650P | 3.343 ms | 0 - 70 MB | NPU
137
+ | ResNet50 | TFLITE | float | Qualcomm® SA8255P | 3.343 ms | 0 - 70 MB | NPU
138
+ | ResNet50 | TFLITE | float | Qualcomm® QCS8450 | 3.727 ms | 0 - 96 MB | NPU
139
+ | ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 3.413 ms | 0 - 52 MB | NPU
140
+ | ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.083 ms | 0 - 67 MB | NPU
 
141
  | ResNet50 | TFLITE | float | Qualcomm® SA7255P | 10.663 ms | 0 - 69 MB | NPU
142
+ | ResNet50 | TFLITE | float | Snapdragon® 8 Elite Mobile | 1.249 ms | 0 - 72 MB | NPU
143
+ | ResNet50 | TFLITE | float | Qualcomm® SA8295P | 3.662 ms | 0 - 50 MB | NPU
144
+ | ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 1.249 ms | 0 - 72 MB | NPU
145
+ | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.571 ms | 0 - 74 MB | NPU
146
+ | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.048 ms | 0 - 81 MB | NPU
147
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 3.358 ms | 0 - 27 MB | NPU
148
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8275 | 1.744 ms | 0 - 42 MB | NPU
149
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.762 ms | 0 - 169 MB | NPU
150
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.959 ms | 0 - 45 MB | NPU
151
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8650P | 0.959 ms | 0 - 45 MB | NPU
152
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8255P | 0.959 ms | 0 - 45 MB | NPU
153
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8450 | 1.048 ms | 0 - 81 MB | NPU
154
+ | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.499 ms | 0 - 46 MB | NPU
155
+ | ResNet50 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.094 ms | 0 - 49 MB | NPU
156
+ | ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.449 ms | 0 - 43 MB | NPU
157
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.853 ms | 0 - 27 MB | NPU
158
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 6.232 ms | 0 - 161 MB | NPU
159
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® SA7255P | 1.744 ms | 0 - 42 MB | NPU
160
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® SA8295P | 1.282 ms | 0 - 40 MB | NPU
161
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.094 ms | 0 - 49 MB | NPU
162
+ | ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.499 ms | 0 - 46 MB | NPU
163
 
164
  ## License
165
  * The license for the original implementation of ResNet50 can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.57.3",
3
  "precisions": {
4
  "w8a8": {
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/resnet50/releases/v0.57.3/resnet50-tflite-w8a8.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/resnet50/releases/v0.57.3/resnet50-qnn_dlc-w8a8.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/resnet50/releases/v0.57.3/resnet50-onnx-w8a8.zip"
25
  }
26
  }
27
  },
@@ -32,20 +32,20 @@
32
  "qairt": "2.45.0.260326154327",
33
  "litert": "1.4.4"
34
  },
35
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-tflite-float.zip"
36
  },
37
  "qnn_dlc": {
38
  "tool_versions": {
39
  "qairt": "2.45.0.260326154327"
40
  },
41
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-qnn_dlc-float.zip"
42
  },
43
  "onnx": {
44
  "tool_versions": {
45
  "qairt": "2.45.0.260326154327",
46
  "onnx_runtime": "1.25.0"
47
  },
48
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.57.3/resnet50-onnx-float.zip"
49
  }
50
  }
51
  }
 
1
  {
2
+ "version": "0.58.0",
3
  "precisions": {
4
  "w8a8": {
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/resnet50/releases/v0.58.0/resnet50-tflite-w8a8.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/resnet50/releases/v0.58.0/resnet50-qnn_dlc-w8a8.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/resnet50/releases/v0.58.0/resnet50-onnx-w8a8.zip"
25
  }
26
  }
27
  },
 
32
  "qairt": "2.45.0.260326154327",
33
  "litert": "1.4.4"
34
  },
35
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-tflite-float.zip"
36
  },
37
  "qnn_dlc": {
38
  "tool_versions": {
39
  "qairt": "2.45.0.260326154327"
40
  },
41
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-qnn_dlc-float.zip"
42
  },
43
  "onnx": {
44
  "tool_versions": {
45
  "qairt": "2.45.0.260326154327",
46
  "onnx_runtime": "1.25.0"
47
  },
48
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet50/releases/v0.58.0/resnet50-onnx-float.zip"
49
  }
50
  }
51
  }