Hugging Face's logo

duylb54
/
MobileNet-v3-Small

Image Classification
PyTorch
backbone
bu_auto
real_time
android
Model card Files
xet
 Community
duylb54 shreyajn commited on
Commit
3d6d528
·
0 Parent(s):

Duplicate from qualcomm/MobileNet-v3-Small

Browse files

Co-authored-by: Shreya Jain <shreyajn@users.noreply.huggingface.co>

Files changed (4) hide show
  1. .gitattributes +40 -0
  2. LICENSE +1 -0
  3. README.md +134 -0
  4. release_assets.json +39 -0
.gitattributes ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ MobileNet-v3-Small.so filter=lfs diff=lfs merge=lfs -text
37
+ MobileNet-v3-Small_w8a16.so filter=lfs diff=lfs merge=lfs -text
38
+ MobileNet-v3-Small.dlc filter=lfs diff=lfs merge=lfs -text
39
+ DEPLOYMENT_MODEL_LICENSE.pdf filter=lfs diff=lfs merge=lfs -text
40
+ MobileNet-v3-Small_float.dlc filter=lfs diff=lfs merge=lfs -text
LICENSE ADDED
@@ -0,0 +1 @@
 
 
1
+ The license of the original trained model can be found at https://github.com/pytorch/vision/blob/main/LICENSE.
README.md ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: pytorch
3
+ license: other
4
+ tags:
5
+ - backbone
6
+ - bu_auto
7
+ - real_time
8
+ - android
9
+ pipeline_tag: image-classification
10
+
11
+ ---
12
+
13
+ ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/web-assets/model_demo.png)
14
+
15
+ # MobileNet-v3-Small: Optimized for Qualcomm Devices
16
+
17
+ MobileNetV3Small 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.
18
+
19
+ This is based on the implementation of MobileNet-v3-Small found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py).
20
+ 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/main/src/qai_hub_models/models/mobilenet_v3_small) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
21
+
22
+ 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.
23
+
24
+ ## Getting Started
25
+ There are two ways to deploy this model on your device:
26
+
27
+ ### Option 1: Download Pre-Exported Models
28
+
29
+ Below are pre-exported model assets ready for deployment.
30
+
31
+ | Runtime | Precision | Chipset | SDK Versions | Download |
32
+ |---|---|---|---|---|
33
+ | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-onnx-float.zip)
34
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-qnn_dlc-float.zip)
35
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-qnn_dlc-w8a16.zip)
36
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-tflite-float.zip)
37
+
38
+ For more device-specific assets and performance metrics, visit **[MobileNet-v3-Small on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/mobilenet_v3_small)**.
39
+
40
+
41
+ ### Option 2: Export with Custom Configurations
42
+
43
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/mobilenet_v3_small) Python library to compile and export the model with your own:
44
+ - Custom weights (e.g., fine-tuned checkpoints)
45
+ - Custom input shapes
46
+ - Target device and runtime configurations
47
+
48
+ This option is ideal if you need to customize the model beyond the default configuration provided here.
49
+
50
+ See our repository for [MobileNet-v3-Small on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/mobilenet_v3_small) for usage instructions.
51
+
52
+ ## Model Details
53
+
54
+ **Model Type:** Model_use_case.image_classification
55
+
56
+ **Model Stats:**
57
+ - Model checkpoint: Imagenet
58
+ - Input resolution: 224x224
59
+ - Number of parameters: 2.54M
60
+ - Model size (float): 9.71 MB
61
+
62
+ ## Performance Summary
63
+ | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
64
+ |---|---|---|---|---|---|---
65
+ | MobileNet-v3-Small | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.243 ms | 0 - 34 MB | NPU
66
+ | MobileNet-v3-Small | ONNX | float | Snapdragon® X2 Elite | 0.274 ms | 180 - 180 MB | NPU
67
+ | MobileNet-v3-Small | ONNX | float | Snapdragon® X Elite | 0.553 ms | 148 - 148 MB | NPU
68
+ | MobileNet-v3-Small | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.353 ms | 0 - 52 MB | NPU
69
+ | MobileNet-v3-Small | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.551 ms | 0 - 89 MB | NPU
70
+ | MobileNet-v3-Small | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.284 ms | 0 - 30 MB | NPU
71
+ | MobileNet-v3-Small | ONNX | float | Qualcomm® QCS9075 | 0.771 ms | 0 - 51 MB | NPU
72
+ | MobileNet-v3-Small | ONNX | float | Qualcomm® QCS8750 | 0.284 ms | 0 - 30 MB | NPU
73
+ | MobileNet-v3-Small | ONNX | float | Qualcomm® QCS7181 | 0.553 ms | 148 - 148 MB | NPU
74
+ | MobileNet-v3-Small | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.319 ms | 1 - 34 MB | NPU
75
+ | MobileNet-v3-Small | QNN_DLC | float | Snapdragon® X2 Elite | 0.454 ms | 1 - 1 MB | NPU
76
+ | MobileNet-v3-Small | QNN_DLC | float | Snapdragon® X Elite | 0.963 ms | 1 - 1 MB | NPU
77
+ | MobileNet-v3-Small | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.546 ms | 0 - 43 MB | NPU
78
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® QCS8275 | 2.079 ms | 1 - 28 MB | NPU
79
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.834 ms | 1 - 2 MB | NPU
80
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® SA8775P | 1.1 ms | 1 - 32 MB | NPU
81
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® SA8650P | 1.1 ms | 1 - 32 MB | NPU
82
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® SA8255P | 1.1 ms | 1 - 32 MB | NPU
83
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 1.579 ms | 0 - 46 MB | NPU
84
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® SA7255P | 2.079 ms | 1 - 28 MB | NPU
85
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® SA8295P | 1.446 ms | 0 - 29 MB | NPU
86
+ | MobileNet-v3-Small | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.411 ms | 0 - 33 MB | NPU
87
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® QCS9075 | 0.98 ms | 1 - 3 MB | NPU
88
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® QCS8750 | 0.411 ms | 0 - 33 MB | NPU
89
+ | MobileNet-v3-Small | QNN_DLC | float | Qualcomm® QCS7181 | 0.963 ms | 1 - 1 MB | NPU
90
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.309 ms | 0 - 31 MB | NPU
91
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.435 ms | 0 - 0 MB | NPU
92
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Snapdragon® X Elite | 0.925 ms | 0 - 0 MB | NPU
93
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.557 ms | 0 - 39 MB | NPU
94
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 2.15 ms | 0 - 2 MB | NPU
95
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS8275 | 1.7 ms | 0 - 29 MB | NPU
96
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.79 ms | 0 - 9 MB | NPU
97
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® SA8775P | 0.991 ms | 0 - 29 MB | NPU
98
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® SA8650P | 0.991 ms | 0 - 29 MB | NPU
99
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® SA8255P | 0.991 ms | 0 - 29 MB | NPU
100
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 2.816 ms | 0 - 141 MB | NPU
101
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® SA7255P | 1.7 ms | 0 - 29 MB | NPU
102
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® SA8295P | 1.307 ms | 0 - 26 MB | NPU
103
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 0.8 ms | 0 - 26 MB | NPU
104
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.37 ms | 0 - 26 MB | NPU
105
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 0.954 ms | 0 - 2 MB | NPU
106
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 0.988 ms | 0 - 41 MB | NPU
107
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS7790 | 0.8 ms | 0 - 26 MB | NPU
108
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS8750 | 0.37 ms | 0 - 26 MB | NPU
109
+ | MobileNet-v3-Small | QNN_DLC | w8a16 | Qualcomm® QCS7181 | 0.925 ms | 0 - 0 MB | NPU
110
+ | MobileNet-v3-Small | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.321 ms | 0 - 35 MB | NPU
111
+ | MobileNet-v3-Small | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.552 ms | 0 - 44 MB | NPU
112
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® QCS8275 | 2.156 ms | 0 - 29 MB | NPU
113
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.839 ms | 0 - 2 MB | NPU
114
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® SA8775P | 1.141 ms | 0 - 32 MB | NPU
115
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® SA8650P | 1.141 ms | 0 - 32 MB | NPU
116
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® SA8255P | 1.141 ms | 0 - 32 MB | NPU
117
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 1.6 ms | 0 - 46 MB | NPU
118
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® SA7255P | 2.156 ms | 0 - 29 MB | NPU
119
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® SA8295P | 1.485 ms | 0 - 29 MB | NPU
120
+ | MobileNet-v3-Small | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.423 ms | 0 - 34 MB | NPU
121
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® QCS9075 | 1.013 ms | 0 - 8 MB | NPU
122
+ | MobileNet-v3-Small | TFLITE | float | Qualcomm® QCS8750 | 0.423 ms | 0 - 34 MB | NPU
123
+
124
+ ## License
125
+ * The license for the original implementation of MobileNet-v3-Small can be found
126
+ [here](https://github.com/pytorch/vision/blob/main/LICENSE).
127
+
128
+ ## References
129
+ * [Searching for MobileNetV3](https://arxiv.org/abs/1905.02244)
130
+ * [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py)
131
+
132
+ ## Community
133
+ * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
134
+ * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).
release_assets.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "0.55.0",
3
+ "precisions": {
4
+ "w8a16": {
5
+ "universal_assets": {
6
+ "qnn_dlc": {
7
+ "tool_versions": {
8
+ "qairt": "2.45.0.260326154327"
9
+ },
10
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-qnn_dlc-w8a16.zip"
11
+ }
12
+ }
13
+ },
14
+ "float": {
15
+ "universal_assets": {
16
+ "tflite": {
17
+ "tool_versions": {
18
+ "qairt": "2.45.0.260326154327",
19
+ "litert": "1.4.3"
20
+ },
21
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-tflite-float.zip"
22
+ },
23
+ "qnn_dlc": {
24
+ "tool_versions": {
25
+ "qairt": "2.45.0.260326154327"
26
+ },
27
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-qnn_dlc-float.zip"
28
+ },
29
+ "onnx": {
30
+ "tool_versions": {
31
+ "qairt": "2.42.0.251225135753_193295",
32
+ "onnx_runtime": "1.25.0"
33
+ },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mobilenet_v3_small/releases/v0.55.0/mobilenet_v3_small-onnx-float.zip"
35
+ }
36
+ }
37
+ }
38
+ }
39
+ }