Hugging Face's logo

qualcomm
/
RegNet-Y-800MF

Image Classification
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
f1406aa
·
verified ·
1 Parent(s): 1e086bc

v0.61.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.61.0 for changelog.

Files changed (2) hide show
  1. README.md +107 -105
  2. release_assets.json +7 -7
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  library_name: pytorch
3
- license: other
4
  tags:
5
  - android
6
  pipeline_tag: image-classification
@@ -13,7 +13,7 @@ pipeline_tag: image-classification
13
 
14
  RegNet_Y_800MF is part of the RegNet family of models designed for efficient and scalable image classification. It uses a simple yet effective design space to balance performance and computational cost, making it suitable for mobile and edge devices.
15
 
16
- 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/v0.60.0/src/qai_hub_models/models/regnet_y_800mf) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
17
 
18
  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.
19
 
@@ -26,26 +26,26 @@ Below are pre-exported model assets ready for deployment.
26
 
27
  | Runtime | Precision | Chipset | SDK Versions | Download |
28
  |---|---|---|---|---|
29
- | 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/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-onnx-float.zip)
30
- | 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/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-onnx-w8a8.zip)
31
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-qnn_dlc-float.zip)
32
- | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-qnn_dlc-w8a8.zip)
33
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-tflite-float.zip)
34
- | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-tflite-w8a8.zip)
35
 
36
  For more device-specific assets and performance metrics, visit **[RegNet-Y-800MF on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/regnet_y_800mf)**.
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/v0.60.0/src/qai_hub_models/models/regnet_y_800mf) 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 [RegNet-Y-800MF on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/regnet_y_800mf) for usage instructions.
49
 
50
  ## Model Details
51
 
@@ -59,106 +59,108 @@ See our repository for [RegNet-Y-800MF on GitHub](https://github.com/qualcomm/ai
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
62
- | RegNet-Y-800MF | ONNX | float | Snapdragon® X2 Elite | 0.543 ms | 2 - 2 MB | NPU
63
- | RegNet-Y-800MF | ONNX | float | Snapdragon® X Elite | 1.135 ms | 15 - 15 MB | NPU
64
- | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.717 ms | 0 - 77 MB | NPU
65
  | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 1.453 ms | 1 - 76 MB | NPU
66
- | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1.315 ms | 1 - 4 MB | NPU
67
- | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.081 ms | 0 - 18 MB | NPU
68
  | RegNet-Y-800MF | ONNX | float | Qualcomm® QCS8450 | 1.453 ms | 1 - 76 MB | NPU
69
- | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.4 ms | 1 - 3 MB | NPU
70
- | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.135 ms | 15 - 15 MB | NPU
71
- | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.555 ms | 0 - 51 MB | NPU
72
- | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Elite Mobile | 0.555 ms | 0 - 51 MB | NPU
73
- | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.538 ms | 0 - 55 MB | NPU
74
- | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® X2 Elite | 0.376 ms | 1 - 1 MB | NPU
75
- | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® X Elite | 0.718 ms | 7 - 7 MB | NPU
76
- | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.497 ms | 0 - 84 MB | NPU
77
- | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.877 ms | 0 - 84 MB | NPU
78
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.654 ms | 0 - 3 MB | NPU
79
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.705 ms | 0 - 10 MB | NPU
80
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® QCS8450 | 0.877 ms | 0 - 84 MB | NPU
81
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.833 ms | 0 - 3 MB | NPU
82
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.718 ms | 7 - 7 MB | NPU
83
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 2.578 ms | 0 - 62 MB | NPU
84
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.694 ms | 0 - 56 MB | NPU
85
- | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.432 ms | 0 - 56 MB | NPU
86
- | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.432 ms | 0 - 56 MB | NPU
87
- | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.381 ms | 0 - 59 MB | NPU
88
- | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.694 ms | 0 - 56 MB | NPU
89
- | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® X2 Elite | 0.825 ms | 1 - 1 MB | NPU
90
- | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® X Elite | 1.606 ms | 1 - 1 MB | NPU
91
- | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.916 ms | 0 - 75 MB | NPU
92
- | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 2.245 ms | 0 - 71 MB | NPU
93
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1.558 ms | 1 - 4 MB | NPU
94
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 4.011 ms | 0 - 49 MB | NPU
95
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.398 ms | 1 - 2 MB | NPU
96
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8775P | 1.876 ms | 1 - 52 MB | NPU
97
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8650P | 1.876 ms | 1 - 52 MB | NPU
98
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8255P | 1.876 ms | 1 - 52 MB | NPU
99
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® QCS8450 | 2.245 ms | 0 - 71 MB | NPU
100
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.701 ms | 1 - 3 MB | NPU
101
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.606 ms | 1 - 1 MB | NPU
102
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.697 ms | 1 - 51 MB | NPU
103
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA7255P | 4.011 ms | 0 - 49 MB | NPU
104
- | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8295P | 2.128 ms | 0 - 45 MB | NPU
105
- | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.697 ms | 1 - 51 MB | NPU
106
- | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.62 ms | 1 - 56 MB | NPU
107
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.462 ms | 0 - 0 MB | NPU
108
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.84 ms | 0 - 0 MB | NPU
 
109
  | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.496 ms | 0 - 66 MB | NPU
110
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.863 ms | 0 - 70 MB | NPU
111
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.91 ms | 2 - 4 MB | NPU
112
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 0.754 ms | 0 - 3 MB | NPU
113
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.587 ms | 0 - 48 MB | NPU
114
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.693 ms | 0 - 2 MB | NPU
115
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.924 ms | 0 - 49 MB | NPU
116
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA8650P | 0.924 ms | 0 - 49 MB | NPU
117
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA8255P | 0.924 ms | 0 - 49 MB | NPU
118
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 0.863 ms | 0 - 70 MB | NPU
119
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.782 ms | 0 - 2 MB | NPU
120
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.84 ms | 0 - 0 MB | NPU
121
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 2.877 ms | 0 - 48 MB | NPU
122
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.721 ms | 0 - 47 MB | NPU
123
  | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.378 ms | 0 - 51 MB | NPU
124
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.587 ms | 0 - 48 MB | NPU
 
125
  | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.378 ms | 0 - 51 MB | NPU
126
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.314 ms | 0 - 50 MB | NPU
127
- | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.721 ms | 0 - 47 MB | NPU
128
- | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.912 ms | 0 - 88 MB | NPU
129
- | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 2.207 ms | 0 - 79 MB | NPU
130
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1.567 ms | 0 - 18 MB | NPU
131
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.008 ms | 0 - 58 MB | NPU
132
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.396 ms | 0 - 20 MB | NPU
133
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8775P | 1.912 ms | 0 - 62 MB | NPU
134
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8650P | 1.912 ms | 0 - 62 MB | NPU
135
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8255P | 1.912 ms | 0 - 62 MB | NPU
136
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® QCS8450 | 2.207 ms | 0 - 79 MB | NPU
137
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.713 ms | 0 - 17 MB | NPU
138
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.69 ms | 0 - 63 MB | NPU
139
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA7255P | 4.008 ms | 0 - 58 MB | NPU
140
- | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8295P | 2.135 ms | 0 - 50 MB | NPU
141
- | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.69 ms | 0 - 63 MB | NPU
142
- | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.614 ms | 0 - 59 MB | NPU
143
- | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.351 ms | 0 - 66 MB | NPU
144
- | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.678 ms | 0 - 72 MB | NPU
145
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.562 ms | 0 - 9 MB | NPU
146
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 0.575 ms | 0 - 10 MB | NPU
147
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.256 ms | 0 - 48 MB | NPU
148
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.502 ms | 0 - 12 MB | NPU
149
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8775P | 0.765 ms | 0 - 52 MB | NPU
150
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8650P | 0.765 ms | 0 - 52 MB | NPU
151
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8255P | 0.765 ms | 0 - 52 MB | NPU
152
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® QCS8450 | 0.678 ms | 0 - 72 MB | NPU
153
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.635 ms | 0 - 10 MB | NPU
154
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 2.301 ms | 0 - 44 MB | NPU
155
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.537 ms | 0 - 43 MB | NPU
156
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.304 ms | 0 - 52 MB | NPU
157
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA7255P | 1.256 ms | 0 - 48 MB | NPU
158
- | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8295P | 0.9 ms | 0 - 44 MB | NPU
159
- | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.304 ms | 0 - 52 MB | NPU
160
- | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.255 ms | 0 - 44 MB | NPU
161
- | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.537 ms | 0 - 43 MB | NPU
162
 
163
  ## License
164
  * The license for the original implementation of RegNet-Y-800MF can be found
 
1
  ---
2
  library_name: pytorch
3
+ license: bsd-3-clause
4
  tags:
5
  - android
6
  pipeline_tag: image-classification
 
13
 
14
  RegNet_Y_800MF is part of the RegNet family of models designed for efficient and scalable image classification. It uses a simple yet effective design space to balance performance and computational cost, making it suitable for mobile and edge devices.
15
 
16
+ 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/v0.61.0/src/qai_hub_models/models/regnet_y_800mf) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
17
 
18
  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.
19
 
 
26
 
27
  | Runtime | Precision | Chipset | SDK Versions | Download |
28
  |---|---|---|---|---|
29
+ | 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/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-onnx-float.zip)
30
+ | 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/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-onnx-w8a8.zip)
31
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-qnn_dlc-float.zip)
32
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-qnn_dlc-w8a8.zip)
33
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-tflite-float.zip)
34
+ | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-tflite-w8a8.zip)
35
 
36
  For more device-specific assets and performance metrics, visit **[RegNet-Y-800MF on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/regnet_y_800mf)**.
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/v0.61.0/src/qai_hub_models/models/regnet_y_800mf) 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 [RegNet-Y-800MF on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/regnet_y_800mf) for usage instructions.
49
 
50
  ## Model Details
51
 
 
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
62
+ | RegNet-Y-800MF | ONNX | float | Snapdragon® X2 Elite | 0.539 ms | 2 - 2 MB | NPU
63
+ | RegNet-Y-800MF | ONNX | float | Snapdragon® X Elite | 1.143 ms | 14 - 14 MB | NPU
64
+ | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.713 ms | 0 - 78 MB | NPU
65
  | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 1.453 ms | 1 - 76 MB | NPU
66
+ | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1.313 ms | 1 - 4 MB | NPU
67
+ | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.095 ms | 0 - 17 MB | NPU
68
  | RegNet-Y-800MF | ONNX | float | Qualcomm® QCS8450 | 1.453 ms | 1 - 76 MB | NPU
69
+ | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.41 ms | 1 - 3 MB | NPU
70
+ | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.143 ms | 14 - 14 MB | NPU
71
+ | RegNet-Y-800MF | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.558 ms | 0 - 55 MB | NPU
72
+ | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Elite Mobile | 0.558 ms | 0 - 55 MB | NPU
73
+ | RegNet-Y-800MF | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.536 ms | 0 - 55 MB | NPU
74
+ | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® X2 Elite | 0.377 ms | 1 - 1 MB | NPU
75
+ | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® X Elite | 0.719 ms | 7 - 7 MB | NPU
76
+ | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.486 ms | 0 - 85 MB | NPU
77
+ | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.875 ms | 0 - 84 MB | NPU
78
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.644 ms | 0 - 3 MB | NPU
79
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 0.779 ms | 0 - 4 MB | NPU
80
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.694 ms | 0 - 45 MB | NPU
81
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® QCS8450 | 0.875 ms | 0 - 84 MB | NPU
82
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.859 ms | 0 - 3 MB | NPU
83
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.719 ms | 7 - 7 MB | NPU
84
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 2.548 ms | 0 - 61 MB | NPU
85
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.69 ms | 0 - 56 MB | NPU
86
+ | RegNet-Y-800MF | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.433 ms | 0 - 60 MB | NPU
87
+ | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.433 ms | 0 - 60 MB | NPU
88
+ | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.382 ms | 0 - 63 MB | NPU
89
+ | RegNet-Y-800MF | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.69 ms | 0 - 56 MB | NPU
90
+ | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® X2 Elite | 0.81 ms | 1 - 1 MB | NPU
91
+ | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® X Elite | 1.622 ms | 1 - 1 MB | NPU
92
+ | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.917 ms | 0 - 75 MB | NPU
93
+ | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 2.238 ms | 0 - 69 MB | NPU
94
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1.551 ms | 1 - 4 MB | NPU
95
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 4.035 ms | 1 - 50 MB | NPU
96
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.407 ms | 0 - 14 MB | NPU
97
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8775P | 1.873 ms | 1 - 53 MB | NPU
98
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8650P | 1.873 ms | 1 - 53 MB | NPU
99
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8255P | 1.873 ms | 1 - 53 MB | NPU
100
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® QCS8450 | 2.238 ms | 0 - 69 MB | NPU
101
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.702 ms | 1 - 3 MB | NPU
102
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.622 ms | 1 - 1 MB | NPU
103
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.699 ms | 0 - 55 MB | NPU
104
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA7255P | 4.035 ms | 1 - 50 MB | NPU
105
+ | RegNet-Y-800MF | QNN_DLC | float | Qualcomm® SA8295P | 2.167 ms | 0 - 45 MB | NPU
106
+ | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.699 ms | 0 - 55 MB | NPU
107
+ | RegNet-Y-800MF | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.622 ms | 1 - 56 MB | NPU
108
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.443 ms | 0 - 0 MB | NPU
109
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.841 ms | 0 - 0 MB | NPU
110
  | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.496 ms | 0 - 66 MB | NPU
111
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.858 ms | 0 - 71 MB | NPU
112
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.921 ms | 0 - 2 MB | NPU
113
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 0.76 ms | 0 - 3 MB | NPU
114
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.572 ms | 0 - 48 MB | NPU
115
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.696 ms | 0 - 39 MB | NPU
116
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.912 ms | 0 - 49 MB | NPU
117
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA8650P | 0.912 ms | 0 - 49 MB | NPU
118
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA8255P | 0.912 ms | 0 - 49 MB | NPU
119
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 0.858 ms | 0 - 71 MB | NPU
120
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.782 ms | 2 - 4 MB | NPU
121
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.841 ms | 0 - 0 MB | NPU
122
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 2.887 ms | 0 - 48 MB | NPU
123
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.725 ms | 0 - 47 MB | NPU
124
  | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.378 ms | 0 - 51 MB | NPU
125
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.572 ms | 0 - 48 MB | NPU
126
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.129 ms | 0 - 46 MB | NPU
127
  | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.378 ms | 0 - 51 MB | NPU
128
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.315 ms | 0 - 48 MB | NPU
129
+ | RegNet-Y-800MF | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.725 ms | 0 - 47 MB | NPU
130
+ | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.914 ms | 4 - 89 MB | NPU
131
+ | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 2.213 ms | 0 - 77 MB | NPU
132
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1.558 ms | 0 - 18 MB | NPU
133
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.029 ms | 0 - 58 MB | NPU
134
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.399 ms | 0 - 4 MB | NPU
135
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8775P | 1.884 ms | 0 - 62 MB | NPU
136
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8650P | 1.884 ms | 0 - 62 MB | NPU
137
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8255P | 1.884 ms | 0 - 62 MB | NPU
138
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® QCS8450 | 2.213 ms | 0 - 77 MB | NPU
139
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.71 ms | 0 - 17 MB | NPU
140
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.689 ms | 0 - 65 MB | NPU
141
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA7255P | 4.029 ms | 0 - 58 MB | NPU
142
+ | RegNet-Y-800MF | TFLITE | float | Qualcomm® SA8295P | 2.155 ms | 0 - 50 MB | NPU
143
+ | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.689 ms | 0 - 65 MB | NPU
144
+ | RegNet-Y-800MF | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.61 ms | 0 - 58 MB | NPU
145
+ | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.354 ms | 0 - 66 MB | NPU
146
+ | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.675 ms | 0 - 72 MB | NPU
147
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.492 ms | 0 - 9 MB | NPU
148
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 0.591 ms | 0 - 10 MB | NPU
149
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.243 ms | 0 - 47 MB | NPU
150
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.507 ms | 0 - 28 MB | NPU
151
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8775P | 0.735 ms | 0 - 51 MB | NPU
152
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8650P | 0.735 ms | 0 - 51 MB | NPU
153
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8255P | 0.735 ms | 0 - 51 MB | NPU
154
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® QCS8450 | 0.675 ms | 0 - 72 MB | NPU
155
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.615 ms | 0 - 10 MB | NPU
156
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 2.321 ms | 0 - 44 MB | NPU
157
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.534 ms | 0 - 43 MB | NPU
158
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.3 ms | 0 - 43 MB | NPU
159
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA7255P | 1.243 ms | 0 - 47 MB | NPU
160
+ | RegNet-Y-800MF | TFLITE | w8a8 | Qualcomm® SA8295P | 0.912 ms | 0 - 44 MB | NPU
161
+ | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.3 ms | 0 - 43 MB | NPU
162
+ | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.256 ms | 0 - 44 MB | NPU
163
+ | RegNet-Y-800MF | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.534 ms | 0 - 43 MB | NPU
164
 
165
  ## License
166
  * The license for the original implementation of RegNet-Y-800MF can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.60.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
@@ -8,19 +8,19 @@
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/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-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/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-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/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-tflite-float.zip"
24
  }
25
  }
26
  },
@@ -31,19 +31,19 @@
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-onnx-w8a8.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-qnn_dlc-w8a8.zip"
41
  },
42
  "tflite": {
43
  "tool_versions": {
44
  "qairt": "2.45.0.260326154327"
45
  },
46
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.60.0/regnet_y_800mf-tflite-w8a8.zip"
47
  }
48
  }
49
  }
 
1
  {
2
+ "version": "0.61.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
 
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/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-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/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-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/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-tflite-float.zip"
24
  }
25
  }
26
  },
 
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-onnx-w8a8.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-qnn_dlc-w8a8.zip"
41
  },
42
  "tflite": {
43
  "tool_versions": {
44
  "qairt": "2.45.0.260326154327"
45
  },
46
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/regnet_y_800mf/releases/v0.61.0/regnet_y_800mf-tflite-w8a8.zip"
47
  }
48
  }
49
  }