Hugging Face's logo

qualcomm
/
EfficientNet-B0

Image Classification
PyTorch
backbone
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
e384e3e
·
verified ·
1 Parent(s): d2a3c75

v0.47.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.47.0 for changelog.

Files changed (1) hide show
  1. README.md +62 -58
README.md CHANGED
@@ -16,7 +16,7 @@ pipeline_tag: image-classification
16
  EfficientNetB0 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 EfficientNet-B0 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py).
19
- This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/efficientnet_b0) 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
 
@@ -29,25 +29,25 @@ Below are pre-exported model assets ready for deployment.
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
- | ONNX | float | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.46.0/efficientnet_b0-onnx-float.zip)
33
- | ONNX | w8a16 | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.46.0/efficientnet_b0-onnx-w8a16.zip)
34
- | QNN_DLC | float | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.46.0/efficientnet_b0-qnn_dlc-float.zip)
35
- | QNN_DLC | w8a16 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.46.0/efficientnet_b0-qnn_dlc-w8a16.zip)
36
- | TFLITE | float | Universal | QAIRT 2.42, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.46.0/efficientnet_b0-tflite-float.zip)
37
 
38
  For more device-specific assets and performance metrics, visit **[EfficientNet-B0 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/efficientnet_b0)**.
39
 
40
 
41
  ### Option 2: Export with Custom Configurations
42
 
43
- Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/efficientnet_b0) 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 [EfficientNet-B0 on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/efficientnet_b0) for usage instructions.
51
 
52
  ## Model Details
53
 
@@ -63,56 +63,60 @@ See our repository for [EfficientNet-B0 on GitHub](https://github.com/quic/ai-hu
63
  ## Performance Summary
64
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
65
  |---|---|---|---|---|---|---
66
- | EfficientNet-B0 | ONNX | float | Snapdragon® X Elite | 1.391 ms | 13 - 13 MB | NPU
67
- | EfficientNet-B0 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.063 ms | 0 - 134 MB | NPU
68
- | EfficientNet-B0 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 1.402 ms | 0 - 24 MB | NPU
69
- | EfficientNet-B0 | ONNX | float | Qualcomm® QCS9075 | 1.877 ms | 1 - 3 MB | NPU
70
- | EfficientNet-B0 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.839 ms | 0 - 108 MB | NPU
71
- | EfficientNet-B0 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.659 ms | 0 - 107 MB | NPU
72
- | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® X Elite | 1.618 ms | 6 - 6 MB | NPU
73
- | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 1.155 ms | 0 - 146 MB | NPU
74
- | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCS6490 | 122.425 ms | 42 - 44 MB | CPU
75
- | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.634 ms | 0 - 10 MB | NPU
76
- | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCS9075 | 1.851 ms | 0 - 3 MB | NPU
77
- | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCM6690 | 66.291 ms | 42 - 51 MB | CPU
78
- | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.816 ms | 0 - 124 MB | NPU
79
- | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 54.16 ms | 49 - 58 MB | CPU
80
- | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.699 ms | 0 - 123 MB | NPU
81
- | EfficientNet-B0 | QNN_DLC | float | Snapdragon® X Elite | 1.756 ms | 1 - 1 MB | NPU
82
- | EfficientNet-B0 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.076 ms | 0 - 63 MB | NPU
83
- | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 4.898 ms | 1 - 38 MB | NPU
84
- | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 1.554 ms | 1 - 2 MB | NPU
85
- | EfficientNet-B0 | QNN_DLC | float | Qualcomm® SA8775P | 8.846 ms | 1 - 39 MB | NPU
86
- | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS9075 | 1.857 ms | 3 - 5 MB | NPU
87
- | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 3.601 ms | 0 - 73 MB | NPU
88
- | EfficientNet-B0 | QNN_DLC | float | Qualcomm® SA7255P | 4.898 ms | 1 - 38 MB | NPU
89
- | EfficientNet-B0 | QNN_DLC | float | Qualcomm® SA8295P | 3.614 ms | 0 - 45 MB | NPU
90
- | EfficientNet-B0 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.806 ms | 1 - 44 MB | NPU
91
- | EfficientNet-B0 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.614 ms | 1 - 44 MB | NPU
92
- | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.896 ms | 0 - 0 MB | NPU
93
- | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 1.126 ms | 0 - 65 MB | NPU
94
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 4.316 ms | 0 - 2 MB | NPU
95
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 3.293 ms | 0 - 45 MB | NPU
96
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.677 ms | 0 - 2 MB | NPU
97
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.944 ms | 0 - 49 MB | NPU
98
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 1.857 ms | 2 - 4 MB | NPU
99
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 6.533 ms | 0 - 163 MB | NPU
100
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 1.977 ms | 0 - 67 MB | NPU
101
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® SA7255P | 3.293 ms | 0 - 45 MB | NPU
102
- | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® SA8295P | 2.358 ms | 0 - 43 MB | NPU
103
- | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.783 ms | 0 - 43 MB | NPU
104
- | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.674 ms | 0 - 47 MB | NPU
105
- | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.639 ms | 0 - 48 MB | NPU
106
- | EfficientNet-B0 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.073 ms | 0 - 77 MB | NPU
107
- | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 4.911 ms | 0 - 46 MB | NPU
108
- | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 1.542 ms | 0 - 131 MB | NPU
109
- | EfficientNet-B0 | TFLITE | float | Qualcomm® SA8775P | 8.913 ms | 0 - 46 MB | NPU
110
- | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS9075 | 1.877 ms | 0 - 16 MB | NPU
111
- | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 3.616 ms | 0 - 80 MB | NPU
112
- | EfficientNet-B0 | TFLITE | float | Qualcomm® SA7255P | 4.911 ms | 0 - 46 MB | NPU
113
- | EfficientNet-B0 | TFLITE | float | Qualcomm® SA8295P | 3.657 ms | 0 - 52 MB | NPU
114
- | EfficientNet-B0 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.811 ms | 0 - 45 MB | NPU
115
- | EfficientNet-B0 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.612 ms | 0 - 50 MB | NPU
 
 
 
 
116
 
117
  ## License
118
  * The license for the original implementation of EfficientNet-B0 can be found
 
16
  EfficientNetB0 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 EfficientNet-B0 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.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/main/qai_hub_models/models/efficientnet_b0) 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 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.47.0/efficientnet_b0-onnx-float.zip)
33
+ | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.47.0/efficientnet_b0-onnx-w8a16.zip)
34
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.47.0/efficientnet_b0-qnn_dlc-float.zip)
35
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.47.0/efficientnet_b0-qnn_dlc-w8a16.zip)
36
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_b0/releases/v0.47.0/efficientnet_b0-tflite-float.zip)
37
 
38
  For more device-specific assets and performance metrics, visit **[EfficientNet-B0 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/efficientnet_b0)**.
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/qai_hub_models/models/efficientnet_b0) 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 [EfficientNet-B0 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/efficientnet_b0) for usage instructions.
51
 
52
  ## Model Details
53
 
 
63
  ## Performance Summary
64
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
65
  |---|---|---|---|---|---|---
66
+ | EfficientNet-B0 | ONNX | float | Snapdragon® X2 Elite | 0.667 ms | 13 - 13 MB | NPU
67
+ | EfficientNet-B0 | ONNX | float | Snapdragon® X Elite | 1.456 ms | 13 - 13 MB | NPU
68
+ | EfficientNet-B0 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.899 ms | 0 - 65 MB | NPU
69
+ | EfficientNet-B0 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 1.267 ms | 0 - 15 MB | NPU
70
+ | EfficientNet-B0 | ONNX | float | Qualcomm® QCS9075 | 1.629 ms | 1 - 3 MB | NPU
71
+ | EfficientNet-B0 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.696 ms | 0 - 37 MB | NPU
72
+ | EfficientNet-B0 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.546 ms | 0 - 43 MB | NPU
73
+ | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® X2 Elite | 0.579 ms | 6 - 6 MB | NPU
74
+ | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® X Elite | 1.636 ms | 6 - 6 MB | NPU
75
+ | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.947 ms | 0 - 85 MB | NPU
76
+ | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCS6490 | 113.227 ms | 43 - 46 MB | CPU
77
+ | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.419 ms | 0 - 10 MB | NPU
78
+ | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCS9075 | 1.621 ms | 0 - 3 MB | NPU
79
+ | EfficientNet-B0 | ONNX | w8a16 | Qualcomm® QCM6690 | 48.896 ms | 41 - 51 MB | CPU
80
+ | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.671 ms | 0 - 61 MB | NPU
81
+ | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 39.043 ms | 42 - 51 MB | CPU
82
+ | EfficientNet-B0 | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.559 ms | 0 - 58 MB | NPU
83
+ | EfficientNet-B0 | QNN_DLC | float | Snapdragon® X2 Elite | 0.879 ms | 1 - 1 MB | NPU
84
+ | EfficientNet-B0 | QNN_DLC | float | Snapdragon® X Elite | 1.766 ms | 1 - 1 MB | NPU
85
+ | EfficientNet-B0 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.078 ms | 0 - 63 MB | NPU
86
+ | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 4.93 ms | 1 - 38 MB | NPU
87
+ | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 1.57 ms | 1 - 2 MB | NPU
88
+ | EfficientNet-B0 | QNN_DLC | float | Qualcomm® SA8775P | 2.051 ms | 0 - 43 MB | NPU
89
+ | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS9075 | 1.868 ms | 1 - 3 MB | NPU
90
+ | EfficientNet-B0 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 3.624 ms | 0 - 74 MB | NPU
91
+ | EfficientNet-B0 | QNN_DLC | float | Qualcomm® SA7255P | 4.93 ms | 1 - 38 MB | NPU
92
+ | EfficientNet-B0 | QNN_DLC | float | Qualcomm® SA8295P | 3.65 ms | 1 - 46 MB | NPU
93
+ | EfficientNet-B0 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.818 ms | 1 - 40 MB | NPU
94
+ | EfficientNet-B0 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.607 ms | 1 - 44 MB | NPU
95
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.816 ms | 0 - 0 MB | NPU
96
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.924 ms | 0 - 0 MB | NPU
97
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 1.147 ms | 0 - 65 MB | NPU
98
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 4.115 ms | 2 - 4 MB | NPU
99
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 3.327 ms | 0 - 45 MB | NPU
100
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.692 ms | 0 - 2 MB | NPU
101
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.969 ms | 0 - 49 MB | NPU
102
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 1.857 ms | 0 - 2 MB | NPU
103
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 6.462 ms | 0 - 163 MB | NPU
104
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 1.951 ms | 0 - 67 MB | NPU
105
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® SA7255P | 3.327 ms | 0 - 45 MB | NPU
106
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Qualcomm® SA8295P | 2.419 ms | 0 - 43 MB | NPU
107
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.785 ms | 0 - 45 MB | NPU
108
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.72 ms | 0 - 46 MB | NPU
109
+ | EfficientNet-B0 | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.642 ms | 0 - 48 MB | NPU
110
+ | EfficientNet-B0 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.077 ms | 0 - 77 MB | NPU
111
+ | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 4.939 ms | 0 - 46 MB | NPU
112
+ | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 1.568 ms | 0 - 6 MB | NPU
113
+ | EfficientNet-B0 | TFLITE | float | Qualcomm® SA8775P | 2.047 ms | 0 - 49 MB | NPU
114
+ | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS9075 | 1.876 ms | 0 - 16 MB | NPU
115
+ | EfficientNet-B0 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 3.634 ms | 0 - 82 MB | NPU
116
+ | EfficientNet-B0 | TFLITE | float | Qualcomm® SA7255P | 4.939 ms | 0 - 46 MB | NPU
117
+ | EfficientNet-B0 | TFLITE | float | Qualcomm® SA8295P | 3.683 ms | 0 - 52 MB | NPU
118
+ | EfficientNet-B0 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.824 ms | 0 - 51 MB | NPU
119
+ | EfficientNet-B0 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.609 ms | 0 - 50 MB | NPU
120
 
121
  ## License
122
  * The license for the original implementation of EfficientNet-B0 can be found