Hugging Face's logo

qualcomm
/
NASNet

Image Classification
PyTorch
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
aa82597
·
verified ·
1 Parent(s): e6d6827

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 +48 -60
  2. release_assets.json +8 -20
README.md CHANGED
@@ -15,7 +15,7 @@ pipeline_tag: image-classification
15
  NASNet is a CNN-based architecture discovered via Neural Architecture Search (NAS) that can classify images from the Imagenet dataset.
16
 
17
  This is based on the implementation of NASNet found [here](https://github.com/huggingface/pytorch-image-models/tree/main).
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/nasnet) 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
 
@@ -28,24 +28,23 @@ Below are pre-exported model assets ready for deployment.
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
- | 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/nasnet/releases/v0.58.0/nasnet-onnx-float.zip)
32
- | ONNX | w8a8_mixed_fp16 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nasnet/releases/v0.58.0/nasnet-onnx-w8a8_mixed_fp16.zip)
33
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nasnet/releases/v0.58.0/nasnet-qnn_dlc-float.zip)
34
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nasnet/releases/v0.58.0/nasnet-tflite-float.zip)
35
 
36
  For more device-specific assets and performance metrics, visit **[NASNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/nasnet)**.
37
 
38
 
39
  ### Option 2: Export with Custom Configurations
40
 
41
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/nasnet) Python library to compile and export the model with your own:
42
  - Custom weights (e.g., fine-tuned checkpoints)
43
  - Custom input shapes
44
  - Target device and runtime configurations
45
 
46
  This option is ideal if you need to customize the model beyond the default configuration provided here.
47
 
48
- See our repository for [NASNet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/nasnet) for usage instructions.
49
 
50
  ## Model Details
51
 
@@ -62,59 +61,48 @@ See our repository for [NASNet on GitHub](https://github.com/qualcomm/ai-hub-mod
62
  ## Performance Summary
63
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
64
  |---|---|---|---|---|---|---
65
- | NASNet | ONNX | float | Snapdragon® X2 Elite | 9.577 ms | 1 - 1 MB | NPU
66
- | NASNet | ONNX | float | Snapdragon® X Elite | 18.942 ms | 188 - 188 MB | NPU
67
- | NASNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 14.232 ms | 1 - 571 MB | NPU
68
- | NASNet | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 52.77 ms | 2 - 518 MB | NPU
69
- | NASNet | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 18.818 ms | 0 - 194 MB | NPU
70
- | NASNet | ONNX | float | Qualcomm® QCS8450 | 52.77 ms | 2 - 518 MB | NPU
71
- | NASNet | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 27.975 ms | 1 - 5 MB | NPU
72
- | NASNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 9.504 ms | 1 - 416 MB | NPU
73
- | NASNet | ONNX | float | Snapdragon® 8 Elite Mobile | 11.844 ms | 1 - 405 MB | NPU
74
- | NASNet | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 11.844 ms | 1 - 405 MB | NPU
75
- | NASNet | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 18.942 ms | 188 - 188 MB | NPU
76
- | NASNet | ONNX | w8a8_mixed_fp16 | Snapdragon® X2 Elite | 4.672 ms | 1 - 1 MB | NPU
77
- | NASNet | ONNX | w8a8_mixed_fp16 | Snapdragon® X Elite | 9.409 ms | 99 - 99 MB | NPU
78
- | NASNet | ONNX | w8a8_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 6.52 ms | 0 - 618 MB | NPU
79
- | NASNet | ONNX | w8a8_mixed_fp16 | Snapdragon® 8 Gen 1 Mobile | 19.308 ms | 0 - 598 MB | NPU
80
- | NASNet | ONNX | w8a8_mixed_fp16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 9.23 ms | 1 - 4 MB | NPU
81
- | NASNet | ONNX | w8a8_mixed_fp16 | Qualcomm® QCS8450 | 19.308 ms | 0 - 598 MB | NPU
82
- | NASNet | ONNX | w8a8_mixed_fp16 | Qualcomm® Dragonwing™ IQ-9075 | 9.466 ms | 0 - 3 MB | NPU
83
- | NASNet | ONNX | w8a8_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 4.717 ms | 0 - 482 MB | NPU
84
- | NASNet | ONNX | w8a8_mixed_fp16 | Snapdragon® 8 Elite Mobile | 5.338 ms | 0 - 488 MB | NPU
85
- | NASNet | ONNX | w8a8_mixed_fp16 | Qualcomm® Dragonwing™ Q-8750 | 5.338 ms | 0 - 488 MB | NPU
86
- | NASNet | ONNX | w8a8_mixed_fp16 | Qualcomm® Dragonwing™ IQ-X7181 | 9.409 ms | 99 - 99 MB | NPU
87
- | NASNet | QNN_DLC | float | Snapdragon® X2 Elite | 8.9 ms | 1 - 1 MB | NPU
88
- | NASNet | QNN_DLC | float | Snapdragon® X Elite | 16.762 ms | 1 - 1 MB | NPU
89
- | NASNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 12.288 ms | 1 - 516 MB | NPU
90
- | NASNet | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 48.631 ms | 0 - 474 MB | NPU
91
- | NASNet | QNN_DLC | float | Qualcomm® QCS8275 | 89.179 ms | 2 - 351 MB | NPU
92
- | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 16.416 ms | 1 - 3 MB | NPU
93
- | NASNet | QNN_DLC | float | Qualcomm® SA8775P | 24.423 ms | 1 - 354 MB | NPU
94
- | NASNet | QNN_DLC | float | Qualcomm® SA8650P | 24.423 ms | 1 - 354 MB | NPU
95
- | NASNet | QNN_DLC | float | Qualcomm® SA8255P | 24.423 ms | 1 - 354 MB | NPU
96
- | NASNet | QNN_DLC | float | Qualcomm® QCS8450 | 48.631 ms | 0 - 474 MB | NPU
97
- | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 24.184 ms | 1 - 4 MB | NPU
98
- | NASNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.823 ms | 0 - 367 MB | NPU
99
- | NASNet | QNN_DLC | float | Qualcomm® SA7255P | 89.179 ms | 2 - 351 MB | NPU
100
- | NASNet | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 9.94 ms | 1 - 352 MB | NPU
101
- | NASNet | QNN_DLC | float | Qualcomm® SA8295P | 38.611 ms | 1 - 310 MB | NPU
102
- | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 9.94 ms | 1 - 352 MB | NPU
103
- | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 16.762 ms | 1 - 1 MB | NPU
104
- | NASNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 12.568 ms | 0 - 691 MB | NPU
105
- | NASNet | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 50.677 ms | 0 - 653 MB | NPU
106
- | NASNet | TFLITE | float | Qualcomm® QCS8275 | 89.649 ms | 0 - 538 MB | NPU
107
- | NASNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 16.559 ms | 7 - 10 MB | NPU
108
- | NASNet | TFLITE | float | Qualcomm® SA8775P | 24.741 ms | 0 - 538 MB | NPU
109
- | NASNet | TFLITE | float | Qualcomm® SA8650P | 24.741 ms | 0 - 538 MB | NPU
110
- | NASNet | TFLITE | float | Qualcomm® SA8255P | 24.741 ms | 0 - 538 MB | NPU
111
- | NASNet | TFLITE | float | Qualcomm® QCS8450 | 50.677 ms | 0 - 653 MB | NPU
112
- | NASNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 25.073 ms | 0 - 192 MB | NPU
113
- | NASNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.24 ms | 0 - 550 MB | NPU
114
- | NASNet | TFLITE | float | Qualcomm® SA7255P | 89.649 ms | 0 - 538 MB | NPU
115
- | NASNet | TFLITE | float | Snapdragon® 8 Elite Mobile | 10.325 ms | 0 - 532 MB | NPU
116
- | NASNet | TFLITE | float | Qualcomm® SA8295P | 40.074 ms | 0 - 486 MB | NPU
117
- | NASNet | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 10.325 ms | 0 - 532 MB | NPU
118
 
119
  ## License
120
  * The license for the original implementation of NASNet can be found
 
15
  NASNet is a CNN-based architecture discovered via Neural Architecture Search (NAS) that can classify images from the Imagenet dataset.
16
 
17
  This is based on the implementation of NASNet found [here](https://github.com/huggingface/pytorch-image-models/tree/main).
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/nasnet) 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/nasnet/releases/v0.59.0/nasnet-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/nasnet/releases/v0.59.0/nasnet-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/nasnet/releases/v0.59.0/nasnet-tflite-float.zip)
 
34
 
35
  For more device-specific assets and performance metrics, visit **[NASNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/nasnet)**.
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/nasnet) 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 [NASNet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/nasnet) for usage instructions.
48
 
49
  ## Model Details
50
 
 
61
  ## Performance Summary
62
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
63
  |---|---|---|---|---|---|---
64
+ | NASNet | ONNX | float | Snapdragon® X2 Elite | 9.635 ms | 1 - 1 MB | NPU
65
+ | NASNet | ONNX | float | Snapdragon® X Elite | 18.905 ms | 189 - 189 MB | NPU
66
+ | NASNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 14.121 ms | 1 - 571 MB | NPU
67
+ | NASNet | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 52.409 ms | 2 - 519 MB | NPU
68
+ | NASNet | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 18.849 ms | 0 - 194 MB | NPU
69
+ | NASNet | ONNX | float | Qualcomm® QCS8450 | 52.409 ms | 2 - 519 MB | NPU
70
+ | NASNet | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 27.065 ms | 1 - 5 MB | NPU
71
+ | NASNet | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 18.905 ms | 189 - 189 MB | NPU
72
+ | NASNet | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 11.649 ms | 1 - 405 MB | NPU
73
+ | NASNet | ONNX | float | Snapdragon® 8 Elite Mobile | 11.649 ms | 1 - 405 MB | NPU
74
+ | NASNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 9.506 ms | 0 - 415 MB | NPU
75
+ | NASNet | QNN_DLC | float | Snapdragon® X2 Elite | 8.868 ms | 1 - 1 MB | NPU
76
+ | NASNet | QNN_DLC | float | Snapdragon® X Elite | 16.828 ms | 1 - 1 MB | NPU
77
+ | NASNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 12.314 ms | 0 - 516 MB | NPU
78
+ | NASNet | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 48.435 ms | 0 - 475 MB | NPU
79
+ | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 89.202 ms | 2 - 351 MB | NPU
80
+ | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 16.219 ms | 1 - 4 MB | NPU
81
+ | NASNet | QNN_DLC | float | Qualcomm® SA8775P | 24.393 ms | 1 - 354 MB | NPU
82
+ | NASNet | QNN_DLC | float | Qualcomm® SA8650P | 24.393 ms | 1 - 354 MB | NPU
83
+ | NASNet | QNN_DLC | float | Qualcomm® SA8255P | 24.393 ms | 1 - 354 MB | NPU
84
+ | NASNet | QNN_DLC | float | Qualcomm® QCS8450 | 48.435 ms | 0 - 475 MB | NPU
85
+ | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 24.257 ms | 1 - 4 MB | NPU
86
+ | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 16.828 ms | 1 - 1 MB | NPU
87
+ | NASNet | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 9.824 ms | 1 - 352 MB | NPU
88
+ | NASNet | QNN_DLC | float | Qualcomm® SA7255P | 89.202 ms | 2 - 351 MB | NPU
89
+ | NASNet | QNN_DLC | float | Qualcomm® SA8295P | 38.659 ms | 1 - 310 MB | NPU
90
+ | NASNet | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 9.824 ms | 1 - 352 MB | NPU
91
+ | NASNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.816 ms | 0 - 370 MB | NPU
92
+ | NASNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 12.71 ms | 0 - 695 MB | NPU
93
+ | NASNet | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 49.651 ms | 0 - 653 MB | NPU
94
+ | NASNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 89.633 ms | 0 - 539 MB | NPU
95
+ | NASNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 16.666 ms | 0 - 3 MB | NPU
96
+ | NASNet | TFLITE | float | Qualcomm® SA8775P | 24.747 ms | 0 - 537 MB | NPU
97
+ | NASNet | TFLITE | float | Qualcomm® SA8650P | 24.747 ms | 0 - 537 MB | NPU
98
+ | NASNet | TFLITE | float | Qualcomm® SA8255P | 24.747 ms | 0 - 537 MB | NPU
99
+ | NASNet | TFLITE | float | Qualcomm® QCS8450 | 49.651 ms | 0 - 653 MB | NPU
100
+ | NASNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 25.177 ms | 0 - 192 MB | NPU
101
+ | NASNet | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 10.295 ms | 0 - 539 MB | NPU
102
+ | NASNet | TFLITE | float | Qualcomm® SA7255P | 89.633 ms | 0 - 539 MB | NPU
103
+ | NASNet | TFLITE | float | Qualcomm® SA8295P | 40.084 ms | 0 - 486 MB | NPU
104
+ | NASNet | TFLITE | float | Snapdragon® 8 Elite Mobile | 10.295 ms | 0 - 539 MB | NPU
105
+ | NASNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 8.229 ms | 0 - 549 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
106
 
107
  ## License
108
  * The license for the original implementation of NASNet can be found
release_assets.json CHANGED
@@ -1,38 +1,26 @@
1
  {
2
- "version": "0.58.0",
3
  "precisions": {
4
- "w8a8_mixed_fp16": {
5
- "universal_assets": {
6
- "onnx": {
7
- "tool_versions": {
8
- "qairt": "2.45.0.260326154327",
9
- "onnx_runtime": "1.25.0"
10
- },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nasnet/releases/v0.58.0/nasnet-onnx-w8a8_mixed_fp16.zip"
12
- }
13
- }
14
- },
15
  "float": {
16
  "universal_assets": {
17
- "tflite": {
18
  "tool_versions": {
19
  "qairt": "2.45.0.260326154327",
20
- "litert": "1.4.4"
21
  },
22
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nasnet/releases/v0.58.0/nasnet-tflite-float.zip"
23
  },
24
  "qnn_dlc": {
25
  "tool_versions": {
26
  "qairt": "2.45.0.260326154327"
27
  },
28
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nasnet/releases/v0.58.0/nasnet-qnn_dlc-float.zip"
29
  },
30
- "onnx": {
31
  "tool_versions": {
32
- "qairt": "2.45.0.260326154327",
33
- "onnx_runtime": "1.25.0"
34
  },
35
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nasnet/releases/v0.58.0/nasnet-onnx-float.zip"
36
  }
37
  }
38
  }
 
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/nasnet/releases/v0.59.0/nasnet-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/nasnet/releases/v0.59.0/nasnet-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/nasnet/releases/v0.59.0/nasnet-tflite-float.zip"
24
  }
25
  }
26
  }