Hugging Face's logo

qualcomm
/
EfficientNet-Lite4

Image Classification
PyTorch
backbone
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
dc7c683
·
verified ·
1 Parent(s): 1aec727

v0.64.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.64.0 for changelog.

Files changed (2) hide show
  1. README.md +93 -96
  2. release_assets.json +9 -9
README.md CHANGED
@@ -16,7 +16,7 @@ pipeline_tag: image-classification
16
  EfficientNet-Lite4 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-Lite4 found [here](https://github.com/RangiLyu/EfficientNet-Lite).
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/v0.63.0/src/qai_hub_models/models/efficientnet_lite4) 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
 
@@ -29,26 +29,26 @@ Below are pre-exported model assets ready for deployment.
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
- | ONNX | float | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-onnx-float.zip)
33
- | ONNX | w8a8 | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-onnx-w8a8.zip)
34
- | QNN_DLC | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-qnn_dlc-float.zip)
35
- | QNN_DLC | w8a8 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-qnn_dlc-w8a8.zip)
36
- | TFLITE | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-tflite-float.zip)
37
- | TFLITE | w8a8 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-tflite-w8a8.zip)
38
 
39
  For more device-specific assets and performance metrics, visit **[EfficientNet-Lite4 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/efficientnet_lite4)**.
40
 
41
 
42
  ### Option 2: Export with Custom Configurations
43
 
44
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.63.0/src/qai_hub_models/models/efficientnet_lite4) Python library to compile and export the model with your own:
45
  - Custom weights (e.g., fine-tuned checkpoints)
46
  - Custom input shapes
47
  - Target device and runtime configurations
48
 
49
  This option is ideal if you need to customize the model beyond the default configuration provided here.
50
 
51
- See our repository for [EfficientNet-Lite4 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.63.0/src/qai_hub_models/models/efficientnet_lite4) for usage instructions.
52
 
53
  ## Model Details
54
 
@@ -63,98 +63,95 @@ See our repository for [EfficientNet-Lite4 on GitHub](https://github.com/qualcom
63
  ## Performance Summary
64
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
65
  |---|---|---|---|---|---|---
66
- | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.139 ms | 0 - 61 MB | NPU
67
- | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.502 ms | 0 - 58 MB | NPU
68
- | EfficientNet-Lite4 | ONNX | float | Snapdragon® X2 Elite | 1.423 ms | 1 - 1 MB | NPU
69
- | EfficientNet-Lite4 | ONNX | float | Snapdragon® X Elite | 2.754 ms | 29 - 29 MB | NPU
70
- | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.916 ms | 0 - 104 MB | NPU
71
- | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.325 ms | 1 - 104 MB | NPU
72
- | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 4.619 ms | 1 - 5 MB | NPU
73
- | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2.635 ms | 0 - 49 MB | NPU
74
- | EfficientNet-Lite4 | ONNX | float | Qualcomm® QCS8450 | 5.325 ms | 1 - 104 MB | NPU
75
- | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 4.139 ms | 1 - 4 MB | NPU
76
- | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 2.754 ms | 29 - 29 MB | NPU
77
- | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 1.502 ms | 0 - 58 MB | NPU
78
- | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.6 ms | 0 - 79 MB | NPU
79
- | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.701 ms | 0 - 79 MB | NPU
80
- | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.612 ms | 1 - 1 MB | NPU
81
- | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® X Elite | 1.267 ms | 15 - 15 MB | NPU
82
- | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.866 ms | 0 - 119 MB | NPU
83
- | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.775 ms | 2 - 123 MB | NPU
84
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 5.222 ms | 0 - 4 MB | NPU
85
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.318 ms | 0 - 4 MB | NPU
86
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.224 ms | 0 - 18 MB | NPU
87
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® QCS8450 | 1.775 ms | 2 - 123 MB | NPU
88
  | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.416 ms | 0 - 3 MB | NPU
89
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.267 ms | 15 - 15 MB | NPU
90
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 10.727 ms | 0 - 189 MB | NPU
91
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.457 ms | 0 - 76 MB | NPU
92
- | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.701 ms | 0 - 79 MB | NPU
93
- | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.457 ms | 0 - 76 MB | NPU
94
- | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.25 ms | 1 - 62 MB | NPU
95
- | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.688 ms | 1 - 55 MB | NPU
96
- | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® X2 Elite | 1.831 ms | 1 - 1 MB | NPU
97
- | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® X Elite | 3.523 ms | 1 - 1 MB | NPU
98
- | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 2.197 ms | 0 - 102 MB | NPU
99
- | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 6.711 ms | 0 - 100 MB | NPU
100
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 4.851 ms | 1 - 5 MB | NPU
101
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.132 ms | 1 - 38 MB | NPU
102
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® SA8650P | 4.785 ms | 1 - 55 MB | NPU
103
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® SA8255P | 4.785 ms | 1 - 55 MB | NPU
104
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® QCS8450 | 6.711 ms | 0 - 100 MB | NPU
105
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 4.53 ms | 1 - 4 MB | NPU
106
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 3.523 ms | 1 - 1 MB | NPU
107
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 1.688 ms | 1 - 55 MB | NPU
108
- | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® SA8295P | 6.328 ms | 1 - 53 MB | NPU
109
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.513 ms | 0 - 69 MB | NPU
110
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.649 ms | 0 - 67 MB | NPU
111
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.744 ms | 0 - 0 MB | NPU
112
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.466 ms | 0 - 0 MB | NPU
113
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.892 ms | 0 - 100 MB | NPU
114
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.959 ms | 0 - 113 MB | NPU
115
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 6.325 ms | 0 - 3 MB | NPU
116
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.318 ms | 0 - 3 MB | NPU
117
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.275 ms | 0 - 2 MB | NPU
118
  | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® SA8650P | 1.611 ms | 0 - 66 MB | NPU
119
  | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® SA8255P | 1.611 ms | 0 - 66 MB | NPU
120
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 1.959 ms | 0 - 113 MB | NPU
121
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.408 ms | 2 - 5 MB | NPU
122
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.466 ms | 0 - 0 MB | NPU
123
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 11.26 ms | 0 - 183 MB | NPU
124
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.595 ms | 0 - 70 MB | NPU
125
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.649 ms | 0 - 67 MB | NPU
126
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2.398 ms | 0 - 66 MB | NPU
127
- | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.595 ms | 0 - 70 MB | NPU
128
- | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.25 ms | 0 - 73 MB | NPU
129
- | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.685 ms | 0 - 72 MB | NPU
130
- | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.195 ms | 0 - 113 MB | NPU
131
- | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 6.493 ms | 0 - 115 MB | NPU
132
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.841 ms | 0 - 33 MB | NPU
133
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.117 ms | 0 - 6 MB | NPU
134
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA8650P | 4.792 ms | 0 - 66 MB | NPU
135
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA8255P | 4.792 ms | 0 - 66 MB | NPU
136
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® QCS8450 | 6.493 ms | 0 - 115 MB | NPU
137
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 4.524 ms | 0 - 33 MB | NPU
138
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 1.685 ms | 0 - 72 MB | NPU
139
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA7255P | 36.395 ms | 0 - 28 MB | GPU
140
- | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA8295P | 6.326 ms | 0 - 52 MB | NPU
141
- | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.413 ms | 0 - 62 MB | NPU
142
- | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.505 ms | 0 - 67 MB | NPU
143
- | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.642 ms | 0 - 96 MB | NPU
144
- | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.421 ms | 0 - 109 MB | NPU
145
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 4.678 ms | 0 - 31 MB | NPU
146
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.02 ms | 0 - 18 MB | NPU
147
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.91 ms | 0 - 2 MB | NPU
148
  | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA8650P | 1.278 ms | 0 - 63 MB | NPU
149
  | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA8255P | 1.278 ms | 0 - 63 MB | NPU
150
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® QCS8450 | 1.421 ms | 0 - 109 MB | NPU
151
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.071 ms | 0 - 18 MB | NPU
152
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 10.059 ms | 0 - 176 MB | NPU
153
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.25 ms | 0 - 170 MB | NPU
154
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.505 ms | 0 - 67 MB | NPU
155
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA7255P | 38.3 ms | 1 - 27 MB | GPU
156
- | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA8295P | 2.007 ms | 0 - 60 MB | NPU
157
- | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.25 ms | 0 - 170 MB | NPU
158
 
159
  ## License
160
  * The license for the original implementation of EfficientNet-Lite4 can be found
 
16
  EfficientNet-Lite4 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-Lite4 found [here](https://github.com/RangiLyu/EfficientNet-Lite).
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/v0.64.0/src/qai_hub_models/models/efficientnet_lite4) 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 2.50, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-onnx-float.zip)
33
+ | ONNX | w8a8 | Universal | QAIRT 2.50, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-onnx-w8a8.zip)
34
+ | QNN_DLC | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-qnn_dlc-float.zip)
35
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-qnn_dlc-w8a8.zip)
36
+ | TFLITE | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-tflite-float.zip)
37
+ | TFLITE | w8a8 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-tflite-w8a8.zip)
38
 
39
  For more device-specific assets and performance metrics, visit **[EfficientNet-Lite4 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/efficientnet_lite4)**.
40
 
41
 
42
  ### Option 2: Export with Custom Configurations
43
 
44
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.64.0/src/qai_hub_models/models/efficientnet_lite4) Python library to compile and export the model with your own:
45
  - Custom weights (e.g., fine-tuned checkpoints)
46
  - Custom input shapes
47
  - Target device and runtime configurations
48
 
49
  This option is ideal if you need to customize the model beyond the default configuration provided here.
50
 
51
+ See our repository for [EfficientNet-Lite4 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.64.0/src/qai_hub_models/models/efficientnet_lite4) for usage instructions.
52
 
53
  ## Model Details
54
 
 
63
  ## Performance Summary
64
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
65
  |---|---|---|---|---|---|---
66
+ | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.138 ms | 0 - 60 MB | NPU
67
+ | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.503 ms | 0 - 61 MB | NPU
68
+ | EfficientNet-Lite4 | ONNX | float | Snapdragon® X2 Elite | 1.427 ms | 1 - 1 MB | NPU
69
+ | EfficientNet-Lite4 | ONNX | float | Snapdragon® X Elite | 2.744 ms | 29 - 29 MB | NPU
70
+ | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.918 ms | 0 - 108 MB | NPU
71
+ | EfficientNet-Lite4 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.326 ms | 0 - 102 MB | NPU
72
+ | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 4.613 ms | 1 - 5 MB | NPU
73
+ | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2.637 ms | 1 - 3 MB | NPU
74
+ | EfficientNet-Lite4 | ONNX | float | Qualcomm® QCS8450 | 5.326 ms | 0 - 102 MB | NPU
75
+ | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 4.152 ms | 1 - 4 MB | NPU
76
+ | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 2.744 ms | 29 - 29 MB | NPU
77
+ | EfficientNet-Lite4 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 1.503 ms | 0 - 61 MB | NPU
78
+ | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.588 ms | 0 - 78 MB | NPU
79
+ | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.703 ms | 0 - 79 MB | NPU
80
+ | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.607 ms | 1 - 1 MB | NPU
81
+ | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® X Elite | 1.248 ms | 15 - 15 MB | NPU
82
+ | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.874 ms | 0 - 117 MB | NPU
83
+ | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.779 ms | 0 - 123 MB | NPU
84
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 5.244 ms | 0 - 4 MB | NPU
85
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.316 ms | 0 - 4 MB | NPU
86
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.228 ms | 0 - 3 MB | NPU
87
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® QCS8450 | 1.779 ms | 0 - 123 MB | NPU
88
  | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.416 ms | 0 - 3 MB | NPU
89
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.248 ms | 15 - 15 MB | NPU
90
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 10.721 ms | 0 - 190 MB | NPU
91
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.452 ms | 0 - 77 MB | NPU
92
+ | EfficientNet-Lite4 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.703 ms | 0 - 79 MB | NPU
93
+ | EfficientNet-Lite4 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.452 ms | 0 - 77 MB | NPU
94
+ | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.25 ms | 1 - 64 MB | NPU
95
+ | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.692 ms | 0 - 56 MB | NPU
96
+ | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® X2 Elite | 1.842 ms | 1 - 1 MB | NPU
97
+ | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® X Elite | 3.489 ms | 1 - 1 MB | NPU
98
+ | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 2.191 ms | 0 - 101 MB | NPU
99
+ | EfficientNet-Lite4 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 6.699 ms | 0 - 102 MB | NPU
100
+ | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 4.861 ms | 1 - 5 MB | NPU
101
+ | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.144 ms | 1 - 3 MB | NPU
102
+ | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® QCS8450 | 6.699 ms | 0 - 102 MB | NPU
103
+ | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 4.525 ms | 1 - 4 MB | NPU
104
+ | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 3.489 ms | 1 - 1 MB | NPU
105
+ | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 1.692 ms | 0 - 56 MB | NPU
106
+ | EfficientNet-Lite4 | QNN_DLC | float | Qualcomm® SA8295P | 6.327 ms | 1 - 53 MB | NPU
107
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.519 ms | 0 - 69 MB | NPU
108
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.648 ms | 0 - 67 MB | NPU
109
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.714 ms | 0 - 0 MB | NPU
110
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.895 ms | 0 - 101 MB | NPU
111
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.914 ms | 0 - 113 MB | NPU
112
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 6.163 ms | 0 - 3 MB | NPU
113
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.322 ms | 0 - 3 MB | NPU
114
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.281 ms | 0 - 18 MB | NPU
 
 
 
115
  | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® SA8650P | 1.611 ms | 0 - 66 MB | NPU
116
  | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® SA8255P | 1.611 ms | 0 - 66 MB | NPU
117
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 1.914 ms | 0 - 113 MB | NPU
118
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.406 ms | 0 - 3 MB | NPU
119
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 11.262 ms | 0 - 183 MB | NPU
120
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.591 ms | 0 - 70 MB | NPU
121
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.648 ms | 0 - 67 MB | NPU
122
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2.387 ms | 0 - 66 MB | NPU
123
+ | EfficientNet-Lite4 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.591 ms | 0 - 70 MB | NPU
124
+ | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.247 ms | 0 - 73 MB | NPU
125
+ | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.682 ms | 0 - 72 MB | NPU
126
+ | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.21 ms | 0 - 121 MB | NPU
127
+ | EfficientNet-Lite4 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 6.495 ms | 0 - 116 MB | NPU
128
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.834 ms | 0 - 33 MB | NPU
129
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.077 ms | 0 - 4 MB | NPU
130
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA8775P | 16.255 ms | 0 - 32 MB | GPU
131
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA8650P | 16.255 ms | 0 - 32 MB | GPU
132
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA8255P | 16.255 ms | 0 - 32 MB | GPU
133
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® QCS8450 | 6.495 ms | 0 - 116 MB | NPU
134
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 4.522 ms | 0 - 33 MB | NPU
135
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 1.682 ms | 0 - 72 MB | NPU
136
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA7255P | 36.418 ms | 0 - 28 MB | GPU
137
+ | EfficientNet-Lite4 | TFLITE | float | Qualcomm® SA8295P | 6.322 ms | 0 - 52 MB | NPU
138
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.418 ms | 0 - 64 MB | NPU
139
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.508 ms | 0 - 62 MB | NPU
140
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.656 ms | 0 - 96 MB | NPU
141
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.398 ms | 0 - 109 MB | NPU
142
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 4.683 ms | 0 - 31 MB | NPU
143
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.015 ms | 0 - 18 MB | NPU
144
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.899 ms | 0 - 80 MB | NPU
145
  | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA8650P | 1.278 ms | 0 - 63 MB | NPU
146
  | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA8255P | 1.278 ms | 0 - 63 MB | NPU
147
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® QCS8450 | 1.398 ms | 0 - 109 MB | NPU
148
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.052 ms | 0 - 18 MB | NPU
149
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 10.043 ms | 0 - 176 MB | NPU
150
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.293 ms | 0 - 170 MB | NPU
151
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.508 ms | 0 - 62 MB | NPU
152
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA7255P | 37.673 ms | 1 - 27 MB | GPU
153
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Qualcomm® SA8295P | 2.003 ms | 0 - 59 MB | NPU
154
+ | EfficientNet-Lite4 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.293 ms | 0 - 170 MB | NPU
155
 
156
  ## License
157
  * The license for the original implementation of EfficientNet-Lite4 can be found
release_assets.json CHANGED
@@ -1,26 +1,26 @@
1
  {
2
- "version": "0.63.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
  "onnx": {
7
  "tool_versions": {
8
  "qairt": "2.50.0.260828221209",
9
- "onnx_runtime": "1.27.1"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.50.0.260828221209"
16
  },
17
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.50.0.260828221209"
22
  },
23
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-tflite-float.zip"
24
  }
25
  }
26
  },
@@ -29,21 +29,21 @@
29
  "onnx": {
30
  "tool_versions": {
31
  "qairt": "2.50.0.260828221209",
32
- "onnx_runtime": "1.27.1"
33
  },
34
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-onnx-w8a8.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.50.0.260828221209"
39
  },
40
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-qnn_dlc-w8a8.zip"
41
  },
42
  "tflite": {
43
  "tool_versions": {
44
  "qairt": "2.50.0.260828221209"
45
  },
46
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.63.0/efficientnet_lite4-tflite-w8a8.zip"
47
  }
48
  }
49
  }
 
1
  {
2
+ "version": "0.64.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
  "onnx": {
7
  "tool_versions": {
8
  "qairt": "2.50.0.260828221209",
9
+ "onnx_runtime": "1.30.0"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.50.0.260828221209"
16
  },
17
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.50.0.260828221209"
22
  },
23
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-tflite-float.zip"
24
  }
25
  }
26
  },
 
29
  "onnx": {
30
  "tool_versions": {
31
  "qairt": "2.50.0.260828221209",
32
+ "onnx_runtime": "1.30.0"
33
  },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-onnx-w8a8.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.50.0.260828221209"
39
  },
40
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-qnn_dlc-w8a8.zip"
41
  },
42
  "tflite": {
43
  "tool_versions": {
44
  "qairt": "2.50.0.260828221209"
45
  },
46
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientnet_lite4/releases/v0.64.0/efficientnet_lite4-tflite-w8a8.zip"
47
  }
48
  }
49
  }