Hugging Face's logo

qualcomm
/
ResNet101

Image Classification
PyTorch
backbone
bu_auto
android
Model card Files
xet
 Community
ResNet101
File size: 12,493 Bytes
dc2d740
 
be2cfc2
dc2d740
 
c5ae420
dc2d740
6ae6d74
dc2d740
 
 
1ae9413
dc2d740
b232769
2ccd7bb
dc2d740
 
b232769
46fd379
b232769
 
 
 
 
 
 
 
 
 
 
 
46fd379
 
 
 
 
 
b232769
 
 
 
 
 
46fd379
b232769
 
 
 
 
 
46fd379
b232769
 
 
 
 
 
 
 
 
 
 
 
 
 
 
46fd379
 
 
 
 
 
 
 
 
 
 
4d09f22
46fd379
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4d09f22
46fd379
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21dd927
dc2d740
83369cc
 
21dd927
dc2d740
 
 
 
 
82bded5
dc2d740
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
172
---
library_name: pytorch
license: other
tags:
- backbone
- bu_auto
- android
pipeline_tag: image-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet101/web-assets/model_demo.png)

# ResNet101: Optimized for Qualcomm Devices

ResNet101 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 ResNet101 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/resnet101) 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/resnet101/releases/v0.59.0/resnet101-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/resnet101/releases/v0.59.0/resnet101-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/resnet101/releases/v0.59.0/resnet101-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/resnet101/releases/v0.59.0/resnet101-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/resnet101/releases/v0.59.0/resnet101-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet101/releases/v0.59.0/resnet101-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[ResNet101 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/resnet101)**.


### 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/resnet101) 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 [ResNet101 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/resnet101) for usage instructions.

## Model Details

**Model Type:** Model_use_case.image_classification

**Model Stats:**
- Model checkpoint: Imagenet
- Input resolution: 224x224
- Number of parameters: 44.5M
- Model size (float): 170 MB
- Model size (w8a8): 43.9 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| ResNet101 | ONNX | float | Snapdragon® X2 Elite | 1.625 ms | 2 - 2 MB | NPU
| ResNet101 | ONNX | float | Snapdragon® X Elite | 3.069 ms | 87 - 87 MB | NPU
| ResNet101 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 2.265 ms | 0 - 124 MB | NPU
| ResNet101 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.413 ms | 0 - 93 MB | NPU
| ResNet101 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.071 ms | 0 - 98 MB | NPU
| ResNet101 | ONNX | float | Qualcomm® QCS8450 | 5.413 ms | 0 - 93 MB | NPU
| ResNet101 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 5.035 ms | 0 - 4 MB | NPU
| ResNet101 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 3.069 ms | 87 - 87 MB | NPU
| ResNet101 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 1.877 ms | 0 - 68 MB | NPU
| ResNet101 | ONNX | float | Snapdragon® 8 Elite Mobile | 1.877 ms | 0 - 68 MB | NPU
| ResNet101 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.602 ms | 0 - 68 MB | NPU
| ResNet101 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.566 ms | 1 - 1 MB | NPU
| ResNet101 | ONNX | w8a8 | Snapdragon® X Elite | 1.141 ms | 43 - 43 MB | NPU
| ResNet101 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.969 ms | 0 - 144 MB | NPU
| ResNet101 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.814 ms | 0 - 146 MB | NPU
| ResNet101 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 6.002 ms | 0 - 3 MB | NPU
| ResNet101 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.282 ms | 0 - 51 MB | NPU
| ResNet101 | ONNX | w8a8 | Qualcomm® QCS8450 | 1.814 ms | 0 - 146 MB | NPU
| ResNet101 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.386 ms | 0 - 3 MB | NPU
| ResNet101 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.141 ms | 43 - 43 MB | NPU
| ResNet101 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.837 ms | 6 - 88 MB | NPU
| ResNet101 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.837 ms | 6 - 88 MB | NPU
| ResNet101 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.815 ms | 0 - 84 MB | NPU
| ResNet101 | QNN_DLC | float | Snapdragon® X2 Elite | 1.931 ms | 1 - 1 MB | NPU
| ResNet101 | QNN_DLC | float | Snapdragon® X Elite | 3.54 ms | 1 - 1 MB | NPU
| ResNet101 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 2.423 ms | 0 - 122 MB | NPU
| ResNet101 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 5.936 ms | 0 - 90 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 18.344 ms | 1 - 66 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.306 ms | 1 - 239 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® SA8775P | 5.451 ms | 1 - 67 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® SA8650P | 5.451 ms | 1 - 67 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® SA8255P | 5.451 ms | 1 - 67 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® QCS8450 | 5.936 ms | 0 - 90 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 5.132 ms | 3 - 5 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 3.54 ms | 1 - 1 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 1.954 ms | 1 - 67 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® SA7255P | 18.344 ms | 1 - 66 MB | NPU
| ResNet101 | QNN_DLC | float | Qualcomm® SA8295P | 5.63 ms | 0 - 39 MB | NPU
| ResNet101 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 1.954 ms | 1 - 67 MB | NPU
| ResNet101 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.643 ms | 1 - 69 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.643 ms | 0 - 0 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.257 ms | 0 - 0 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.963 ms | 0 - 126 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.784 ms | 0 - 127 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 6.11 ms | 2 - 4 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 3.037 ms | 0 - 75 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.281 ms | 0 - 125 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 1.514 ms | 0 - 78 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® SA8650P | 1.514 ms | 0 - 78 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® SA8255P | 1.514 ms | 0 - 78 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 1.784 ms | 0 - 127 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.392 ms | 0 - 2 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.257 ms | 0 - 0 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 11.833 ms | 0 - 198 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.804 ms | 0 - 85 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.807 ms | 0 - 77 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 3.037 ms | 0 - 75 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.974 ms | 0 - 72 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.807 ms | 0 - 77 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.742 ms | 0 - 77 MB | NPU
| ResNet101 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.804 ms | 0 - 85 MB | NPU
| ResNet101 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.391 ms | 0 - 176 MB | NPU
| ResNet101 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5.914 ms | 0 - 146 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 18.249 ms | 0 - 107 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.308 ms | 0 - 2 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® SA8775P | 5.42 ms | 0 - 112 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® SA8650P | 5.42 ms | 0 - 112 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® SA8255P | 5.42 ms | 0 - 112 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® QCS8450 | 5.914 ms | 0 - 146 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 5.121 ms | 0 - 88 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 1.924 ms | 0 - 112 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® SA7255P | 18.249 ms | 0 - 107 MB | NPU
| ResNet101 | TFLITE | float | Qualcomm® SA8295P | 5.554 ms | 0 - 77 MB | NPU
| ResNet101 | TFLITE | float | Snapdragon® 8 Elite Mobile | 1.924 ms | 0 - 112 MB | NPU
| ResNet101 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.637 ms | 0 - 113 MB | NPU
| ResNet101 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.861 ms | 0 - 134 MB | NPU
| ResNet101 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.631 ms | 0 - 137 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 5.336 ms | 0 - 45 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 2.756 ms | 0 - 71 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.139 ms | 0 - 3 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® SA8775P | 1.396 ms | 0 - 74 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® SA8650P | 1.396 ms | 0 - 74 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® SA8255P | 1.396 ms | 0 - 74 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® QCS8450 | 1.631 ms | 0 - 137 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.326 ms | 0 - 45 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 11.282 ms | 0 - 196 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.67 ms | 0 - 78 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.742 ms | 0 - 75 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® SA7255P | 2.756 ms | 0 - 71 MB | NPU
| ResNet101 | TFLITE | w8a8 | Qualcomm® SA8295P | 1.802 ms | 0 - 68 MB | NPU
| ResNet101 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.742 ms | 0 - 75 MB | NPU
| ResNet101 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.705 ms | 0 - 74 MB | NPU
| ResNet101 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.67 ms | 0 - 78 MB | NPU

## License
* The license for the original implementation of ResNet101 can be found
  [here](https://github.com/pytorch/vision/blob/main/LICENSE).

## References
* [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385)
* [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).