Hugging Face's logo

qualcomm
/
WideResNet50

Image Classification
PyTorch
backbone
android
Model card Files
xet
 Community
WideResNet50
File size: 12,835 Bytes
c5ed3e5
 
b854422
c5ed3e5
 
 
8ae5e2a
c5ed3e5
 
 
28cde4f
c5ed3e5
fb6163e
dd98876
c5ed3e5
 
fb6163e
8f484e9
fb6163e
 
 
 
 
 
 
 
 
 
 
 
8f484e9
 
 
 
 
 
fb6163e
 
 
 
 
 
8f484e9
fb6163e
 
 
 
 
 
8f484e9
fb6163e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8f484e9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cd0f637
8f484e9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3b7145b
c5ed3e5
d2140e7
 
3b7145b
c5ed3e5
 
 
 
 
5c4da85
c5ed3e5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
---
library_name: pytorch
license: other
tags:
- backbone
- android
pipeline_tag: image-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/wideresnet50/web-assets/model_demo.png)

# WideResNet50: Optimized for Qualcomm Devices

WideResNet50 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.

This is based on the implementation of WideResNet50 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py).
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/wideresnet50) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

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.

## Getting Started
There are two ways to deploy this model on your device:

### Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| 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/wideresnet50/releases/v0.59.0/wideresnet50-onnx-float.zip)
| ONNX | w8a8 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/wideresnet50/releases/v0.59.0/wideresnet50-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/wideresnet50/releases/v0.59.0/wideresnet50-qnn_dlc-float.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/wideresnet50/releases/v0.59.0/wideresnet50-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/wideresnet50/releases/v0.59.0/wideresnet50-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/wideresnet50/releases/v0.59.0/wideresnet50-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[WideResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/wideresnet50)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/wideresnet50) Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for [WideResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/wideresnet50) for usage instructions.

## Model Details

**Model Type:** Model_use_case.image_classification

**Model Stats:**
- Model checkpoint: Imagenet
- Input resolution: 224x224
- Number of parameters: 68.9M
- Model size (float): 263 MB
- Model size (w8a8): 66.6 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| WideResNet50 | ONNX | float | Snapdragon® X2 Elite | 2.128 ms | 2 - 2 MB | NPU
| WideResNet50 | ONNX | float | Snapdragon® X Elite | 4.223 ms | 132 - 132 MB | NPU
| WideResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 3.522 ms | 0 - 218 MB | NPU
| WideResNet50 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 8.791 ms | 1 - 200 MB | NPU
| WideResNet50 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.514 ms | 0 - 489 MB | NPU
| WideResNet50 | ONNX | float | Qualcomm® QCS8450 | 8.791 ms | 1 - 200 MB | NPU
| WideResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 6.606 ms | 0 - 4 MB | NPU
| WideResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 4.223 ms | 132 - 132 MB | NPU
| WideResNet50 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 2.856 ms | 0 - 174 MB | NPU
| WideResNet50 | ONNX | float | Snapdragon® 8 Elite Mobile | 2.856 ms | 0 - 174 MB | NPU
| WideResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.427 ms | 0 - 172 MB | NPU
| WideResNet50 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.807 ms | 1 - 1 MB | NPU
| WideResNet50 | ONNX | w8a8 | Snapdragon® X Elite | 1.681 ms | 67 - 67 MB | NPU
| WideResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.401 ms | 0 - 132 MB | NPU
| WideResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 2.596 ms | 0 - 136 MB | NPU
| WideResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 9.073 ms | 0 - 3 MB | NPU
| WideResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.861 ms | 0 - 92 MB | NPU
| WideResNet50 | ONNX | w8a8 | Qualcomm® QCS8450 | 2.596 ms | 0 - 136 MB | NPU
| WideResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.928 ms | 0 - 3 MB | NPU
| WideResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.681 ms | 67 - 67 MB | NPU
| WideResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 1.22 ms | 0 - 58 MB | NPU
| WideResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 1.22 ms | 0 - 58 MB | NPU
| WideResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.158 ms | 0 - 60 MB | NPU
| WideResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 2.534 ms | 1 - 1 MB | NPU
| WideResNet50 | QNN_DLC | float | Snapdragon® X Elite | 4.741 ms | 1 - 1 MB | NPU
| WideResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 3.643 ms | 0 - 224 MB | NPU
| WideResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 9.208 ms | 0 - 198 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 24.503 ms | 1 - 174 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.814 ms | 1 - 297 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 7.142 ms | 1 - 169 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® SA8650P | 7.142 ms | 1 - 169 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® SA8255P | 7.142 ms | 1 - 169 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 9.208 ms | 0 - 198 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 7.118 ms | 1 - 3 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 4.741 ms | 1 - 1 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 2.945 ms | 0 - 174 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 24.503 ms | 1 - 174 MB | NPU
| WideResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 7.833 ms | 1 - 152 MB | NPU
| WideResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 2.945 ms | 0 - 174 MB | NPU
| WideResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.575 ms | 1 - 174 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.883 ms | 0 - 0 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.794 ms | 0 - 0 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.39 ms | 0 - 129 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 2.512 ms | 0 - 132 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 9.407 ms | 0 - 2 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 3.978 ms | 0 - 55 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.852 ms | 0 - 2 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 2.024 ms | 0 - 57 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8650P | 2.024 ms | 0 - 57 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8255P | 2.024 ms | 0 - 57 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 2.512 ms | 0 - 132 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.89 ms | 2 - 4 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.794 ms | 0 - 0 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 21.331 ms | 0 - 216 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 2.852 ms | 0 - 215 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 1.197 ms | 0 - 56 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 3.978 ms | 0 - 55 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2.731 ms | 0 - 53 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 1.197 ms | 0 - 56 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.093 ms | 0 - 59 MB | NPU
| WideResNet50 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.852 ms | 0 - 215 MB | NPU
| WideResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 3.568 ms | 0 - 213 MB | NPU
| WideResNet50 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 8.178 ms | 0 - 193 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 23.772 ms | 0 - 101 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.752 ms | 0 - 2 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® SA8775P | 7.04 ms | 0 - 102 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® SA8650P | 7.04 ms | 0 - 102 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® SA8255P | 7.04 ms | 0 - 102 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® QCS8450 | 8.178 ms | 0 - 193 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 6.749 ms | 0 - 134 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 2.858 ms | 0 - 107 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® SA7255P | 23.772 ms | 0 - 101 MB | NPU
| WideResNet50 | TFLITE | float | Qualcomm® SA8295P | 7.761 ms | 0 - 88 MB | NPU
| WideResNet50 | TFLITE | float | Snapdragon® 8 Elite Mobile | 2.858 ms | 0 - 107 MB | NPU
| WideResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.538 ms | 0 - 103 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.329 ms | 0 - 129 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 2.397 ms | 0 - 129 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 8.978 ms | 0 - 68 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 3.729 ms | 0 - 52 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.733 ms | 0 - 2 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® SA8775P | 1.883 ms | 0 - 54 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® SA8650P | 1.883 ms | 0 - 54 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® SA8255P | 1.883 ms | 0 - 54 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® QCS8450 | 2.397 ms | 0 - 129 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.747 ms | 0 - 68 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 18.207 ms | 0 - 215 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 2.719 ms | 0 - 207 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 1.122 ms | 0 - 55 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® SA7255P | 3.729 ms | 0 - 52 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Qualcomm® SA8295P | 2.578 ms | 0 - 50 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 1.122 ms | 0 - 55 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.052 ms | 0 - 55 MB | NPU
| WideResNet50 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.719 ms | 0 - 207 MB | NPU

## License
* The license for the original implementation of WideResNet50 can be found
  [here](https://github.com/pytorch/vision/blob/main/LICENSE).

## References
* [Wide Residual Networks](https://arxiv.org/abs/1605.07146)
* [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py)

## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).