Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -34,39 +34,39 @@ More details on model performance across various devices, can be found
|
|
34 |
|
35 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
36 |
|---|---|---|---|---|---|---|---|---|
|
37 |
-
| AOT-GAN | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 150.
|
38 |
-
| AOT-GAN | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 153.
|
39 |
-
| AOT-GAN | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 105.
|
40 |
-
| AOT-GAN | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE |
|
41 |
-
| AOT-GAN | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN |
|
42 |
-
| AOT-GAN | Samsung Galaxy S24 | Snapdragon
|
43 |
-
| AOT-GAN | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 118.
|
44 |
-
| AOT-GAN | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN |
|
45 |
-
| AOT-GAN | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX |
|
46 |
-
| AOT-GAN | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 150.
|
47 |
-
| AOT-GAN | QCS8550 (Proxy) | QCS8550 Proxy | QNN |
|
48 |
-
| AOT-GAN | SA7255P ADP | SA7255P | TFLITE | 3625.
|
49 |
-
| AOT-GAN | SA7255P ADP | SA7255P | QNN | 3580.
|
50 |
-
| AOT-GAN | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 150.
|
51 |
-
| AOT-GAN | SA8255 (Proxy) | SA8255P Proxy | QNN | 102.
|
52 |
-
| AOT-GAN | SA8295P ADP | SA8295P | TFLITE | 218.
|
53 |
-
| AOT-GAN | SA8295P ADP | SA8295P | QNN | 164.
|
54 |
-
| AOT-GAN | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 150.
|
55 |
-
| AOT-GAN | SA8650 (Proxy) | SA8650P Proxy | QNN |
|
56 |
-
| AOT-GAN | SA8775P ADP | SA8775P | TFLITE | 241.
|
57 |
-
| AOT-GAN | SA8775P ADP | SA8775P | QNN | 181.
|
58 |
-
| AOT-GAN | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE |
|
59 |
-
| AOT-GAN | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 194.
|
60 |
-
| AOT-GAN | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN |
|
61 |
-
| AOT-GAN | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 102.
|
62 |
|
63 |
|
64 |
|
65 |
|
66 |
## Installation
|
67 |
|
68 |
-
This model can be installed as a Python package via pip.
|
69 |
|
|
|
70 |
```bash
|
71 |
pip install qai-hub-models
|
72 |
```
|
@@ -122,8 +122,8 @@ Profiling Results
|
|
122 |
AOT-GAN
|
123 |
Device : Samsung Galaxy S23 (13)
|
124 |
Runtime : TFLITE
|
125 |
-
Estimated inference time (ms) : 150.
|
126 |
-
Estimated peak memory usage (MB): [3,
|
127 |
Total # Ops : 235
|
128 |
Compute Unit(s) : NPU (235 ops)
|
129 |
```
|
@@ -150,7 +150,7 @@ from qai_hub_models.models.aotgan import Model
|
|
150 |
torch_model = Model.from_pretrained()
|
151 |
|
152 |
# Device
|
153 |
-
device = hub.Device("Samsung Galaxy
|
154 |
|
155 |
# Trace model
|
156 |
input_shape = torch_model.get_input_spec()
|
@@ -242,7 +242,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
|
|
242 |
|
243 |
|
244 |
## License
|
245 |
-
* The license for the original implementation of AOT-GAN can be found
|
|
|
246 |
* 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)
|
247 |
|
248 |
|
|
|
34 |
|
35 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
36 |
|---|---|---|---|---|---|---|---|---|
|
37 |
+
| AOT-GAN | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 150.216 ms | 3 - 38 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
38 |
+
| AOT-GAN | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 153.714 ms | 1 - 32 MB | FP16 | NPU | [AOT-GAN.so](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.so) |
|
39 |
+
| AOT-GAN | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 105.712 ms | 6 - 113 MB | FP16 | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
|
40 |
+
| AOT-GAN | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 112.126 ms | 2 - 60 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
41 |
+
| AOT-GAN | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 112.4 ms | 4 - 62 MB | FP16 | NPU | [AOT-GAN.so](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.so) |
|
42 |
+
| AOT-GAN | Samsung Galaxy S24 | Snapdragon�� 8 Gen 3 | ONNX | 79.635 ms | 12 - 113 MB | FP16 | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
|
43 |
+
| AOT-GAN | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 118.224 ms | 2 - 65 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
44 |
+
| AOT-GAN | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 98.639 ms | 3 - 66 MB | FP16 | NPU | Use Export Script |
|
45 |
+
| AOT-GAN | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 76.142 ms | 11 - 114 MB | FP16 | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
|
46 |
+
| AOT-GAN | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 150.247 ms | 3 - 41 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
47 |
+
| AOT-GAN | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 102.477 ms | 4 - 7 MB | FP16 | NPU | Use Export Script |
|
48 |
+
| AOT-GAN | SA7255P ADP | SA7255P | TFLITE | 3625.101 ms | 0 - 61 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
49 |
+
| AOT-GAN | SA7255P ADP | SA7255P | QNN | 3580.217 ms | 2 - 11 MB | FP16 | NPU | Use Export Script |
|
50 |
+
| AOT-GAN | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 150.364 ms | 3 - 36 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
51 |
+
| AOT-GAN | SA8255 (Proxy) | SA8255P Proxy | QNN | 102.639 ms | 4 - 7 MB | FP16 | NPU | Use Export Script |
|
52 |
+
| AOT-GAN | SA8295P ADP | SA8295P | TFLITE | 218.903 ms | 3 - 52 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
53 |
+
| AOT-GAN | SA8295P ADP | SA8295P | QNN | 164.565 ms | 1 - 16 MB | FP16 | NPU | Use Export Script |
|
54 |
+
| AOT-GAN | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 150.746 ms | 3 - 42 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
55 |
+
| AOT-GAN | SA8650 (Proxy) | SA8650P Proxy | QNN | 104.201 ms | 5 - 7 MB | FP16 | NPU | Use Export Script |
|
56 |
+
| AOT-GAN | SA8775P ADP | SA8775P | TFLITE | 241.948 ms | 3 - 64 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
57 |
+
| AOT-GAN | SA8775P ADP | SA8775P | QNN | 181.669 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
|
58 |
+
| AOT-GAN | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 195.56 ms | 3 - 53 MB | FP16 | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
|
59 |
+
| AOT-GAN | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 194.959 ms | 4 - 50 MB | FP16 | NPU | Use Export Script |
|
60 |
+
| AOT-GAN | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 105.122 ms | 4 - 4 MB | FP16 | NPU | Use Export Script |
|
61 |
+
| AOT-GAN | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 102.363 ms | 30 - 30 MB | FP16 | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
|
62 |
|
63 |
|
64 |
|
65 |
|
66 |
## Installation
|
67 |
|
|
|
68 |
|
69 |
+
Install the package via pip:
|
70 |
```bash
|
71 |
pip install qai-hub-models
|
72 |
```
|
|
|
122 |
AOT-GAN
|
123 |
Device : Samsung Galaxy S23 (13)
|
124 |
Runtime : TFLITE
|
125 |
+
Estimated inference time (ms) : 150.2
|
126 |
+
Estimated peak memory usage (MB): [3, 38]
|
127 |
Total # Ops : 235
|
128 |
Compute Unit(s) : NPU (235 ops)
|
129 |
```
|
|
|
150 |
torch_model = Model.from_pretrained()
|
151 |
|
152 |
# Device
|
153 |
+
device = hub.Device("Samsung Galaxy S24")
|
154 |
|
155 |
# Trace model
|
156 |
input_shape = torch_model.get_input_spec()
|
|
|
242 |
|
243 |
|
244 |
## License
|
245 |
+
* The license for the original implementation of AOT-GAN can be found
|
246 |
+
[here](https://github.com/taki0112/AttnGAN-Tensorflow/blob/master/LICENSE).
|
247 |
* 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)
|
248 |
|
249 |
|