Hugging Face's logo

qualcomm
/
QuickSRNetMedium-Quantized

Image-to-Image
PyTorch
LiteRT
ONNX
quantized
android
Model card Files Community
qaihm-bot commited on
Commit
d8ca7ed
·
verified ·
1 Parent(s): faf7416

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +43 -19
README.md CHANGED
@@ -15,7 +15,7 @@ tags:
15
 
16
  QuickSRNet Medium is designed for upscaling images on mobile platforms to sharpen in real-time.
17
 
18
- This model is an implementation of QuickSRNetMedium-Quantized found [here](https://github.com/quic/aimet-model-zoo/tree/develop/aimet_zoo_torch/quicksrnet).
19
  This repository provides scripts to run QuickSRNetMedium-Quantized on Qualcomm® devices.
20
  More details on model performance across various devices, can be found
21
  [here](https://aihub.qualcomm.com/models/quicksrnetmedium_quantized).
@@ -30,15 +30,35 @@ More details on model performance across various devices, can be found
30
  - Number of parameters: 55.0K
31
  - Model size: 67.2 KB
32
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
 
34
 
35
 
36
- | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
- | ---|---|---|---|---|---|---|---|
38
- | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 1.111 ms | 0 - 1 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite)
39
- | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 0.512 ms | 0 - 64 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.so](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.so)
40
-
41
-
42
 
43
  ## Installation
44
 
@@ -94,16 +114,16 @@ device. This script does the following:
94
  ```bash
95
  python -m qai_hub_models.models.quicksrnetmedium_quantized.export
96
  ```
97
-
98
  ```
99
- Profile Job summary of QuickSRNetMedium-Quantized
100
- --------------------------------------------------
101
- Device: Snapdragon X Elite CRD (11)
102
- Estimated Inference Time: 0.52 ms
103
- Estimated Peak Memory Range: 0.07-0.07 MB
104
- Compute Units: NPU (10) | Total (10)
105
-
106
-
 
107
  ```
108
 
109
 
@@ -142,15 +162,19 @@ provides instructions on how to use the `.so` shared library in an Android appl
142
  Get more details on QuickSRNetMedium-Quantized's performance across various devices [here](https://aihub.qualcomm.com/models/quicksrnetmedium_quantized).
143
  Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
144
 
 
145
  ## License
146
- - The license for the original implementation of QuickSRNetMedium-Quantized can be found
147
- [here](https://github.com/quic/aimet-model-zoo/blob/develop/LICENSE.pdf).
148
- - The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
 
149
 
150
  ## References
151
  * [QuickSRNet: Plain Single-Image Super-Resolution Architecture for Faster Inference on Mobile Platforms](https://arxiv.org/abs/2303.04336)
152
  * [Source Model Implementation](https://github.com/quic/aimet-model-zoo/tree/develop/aimet_zoo_torch/quicksrnet)
153
 
 
 
154
  ## Community
155
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
156
  * For questions or feedback please [reach out to us](mailto:[email protected]).
 
15
 
16
  QuickSRNet Medium is designed for upscaling images on mobile platforms to sharpen in real-time.
17
 
18
+ This model is an implementation of QuickSRNetMedium-Quantized found [here]({source_repo}).
19
  This repository provides scripts to run QuickSRNetMedium-Quantized on Qualcomm® devices.
20
  More details on model performance across various devices, can be found
21
  [here](https://aihub.qualcomm.com/models/quicksrnetmedium_quantized).
 
30
  - Number of parameters: 55.0K
31
  - Model size: 67.2 KB
32
 
33
+ | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
34
+ |---|---|---|---|---|---|---|---|---|
35
+ | QuickSRNetMedium-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.127 ms | 1 - 64 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
36
+ | QuickSRNetMedium-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 0.519 ms | 0 - 4 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.so](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.so) |
37
+ | QuickSRNetMedium-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 0.676 ms | 0 - 1 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.onnx](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.onnx) |
38
+ | QuickSRNetMedium-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 0.899 ms | 0 - 22 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
39
+ | QuickSRNetMedium-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.359 ms | 0 - 11 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.so](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.so) |
40
+ | QuickSRNetMedium-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 0.503 ms | 0 - 23 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.onnx](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.onnx) |
41
+ | QuickSRNetMedium-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 3.558 ms | 2 - 17 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
42
+ | QuickSRNetMedium-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 1.05 ms | 0 - 8 MB | INT8 | NPU | Use Export Script |
43
+ | QuickSRNetMedium-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 12.711 ms | 2 - 7 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
44
+ | QuickSRNetMedium-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.115 ms | 0 - 1 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
45
+ | QuickSRNetMedium-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.412 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
46
+ | QuickSRNetMedium-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.106 ms | 0 - 6 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
47
+ | QuickSRNetMedium-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.413 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
48
+ | QuickSRNetMedium-Quantized | SA8775 (Proxy) | SA8775P Proxy | TFLITE | 1.134 ms | 2 - 68 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
49
+ | QuickSRNetMedium-Quantized | SA8775 (Proxy) | SA8775P Proxy | QNN | 0.416 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
50
+ | QuickSRNetMedium-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.108 ms | 0 - 3 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
51
+ | QuickSRNetMedium-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.415 ms | 0 - 1 MB | INT8 | NPU | Use Export Script |
52
+ | QuickSRNetMedium-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.368 ms | 0 - 24 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
53
+ | QuickSRNetMedium-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 0.581 ms | 0 - 13 MB | INT8 | NPU | Use Export Script |
54
+ | QuickSRNetMedium-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.845 ms | 0 - 15 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.tflite) |
55
+ | QuickSRNetMedium-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.303 ms | 0 - 9 MB | INT8 | NPU | Use Export Script |
56
+ | QuickSRNetMedium-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.373 ms | 0 - 15 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.onnx](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.onnx) |
57
+ | QuickSRNetMedium-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.521 ms | 0 - 0 MB | INT8 | NPU | Use Export Script |
58
+ | QuickSRNetMedium-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 0.777 ms | 3 - 3 MB | INT8 | NPU | [QuickSRNetMedium-Quantized.onnx](https://huggingface.co/qualcomm/QuickSRNetMedium-Quantized/blob/main/QuickSRNetMedium-Quantized.onnx) |
59
 
60
 
61
 
 
 
 
 
 
 
62
 
63
  ## Installation
64
 
 
114
  ```bash
115
  python -m qai_hub_models.models.quicksrnetmedium_quantized.export
116
  ```
 
117
  ```
118
+ Profiling Results
119
+ ------------------------------------------------------------
120
+ QuickSRNetMedium-Quantized
121
+ Device : Samsung Galaxy S23 (13)
122
+ Runtime : TFLITE
123
+ Estimated inference time (ms) : 1.1
124
+ Estimated peak memory usage (MB): [1, 64]
125
+ Total # Ops : 19
126
+ Compute Unit(s) : NPU (16 ops) CPU (3 ops)
127
  ```
128
 
129
 
 
162
  Get more details on QuickSRNetMedium-Quantized's performance across various devices [here](https://aihub.qualcomm.com/models/quicksrnetmedium_quantized).
163
  Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
164
 
165
+
166
  ## License
167
+ * The license for the original implementation of QuickSRNetMedium-Quantized can be found [here](https://github.com/quic/aimet-model-zoo/blob/develop/LICENSE.pdf).
168
+ * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
169
+
170
+
171
 
172
  ## References
173
  * [QuickSRNet: Plain Single-Image Super-Resolution Architecture for Faster Inference on Mobile Platforms](https://arxiv.org/abs/2303.04336)
174
  * [Source Model Implementation](https://github.com/quic/aimet-model-zoo/tree/develop/aimet_zoo_torch/quicksrnet)
175
 
176
+
177
+
178
  ## Community
179
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
180
  * For questions or feedback please [reach out to us](mailto:[email protected]).