Hugging Face's logo

qualcomm
/
AOT-GAN

Image-to-Image
PyTorch
android
Model card Files
xet
 Community
1
qaihm-bot commited on
Commit
07328f3
·
verified ·
1 Parent(s): 42a50b0

v0.30.5

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.30.5 for changelog.

Files changed (1) hide show
  1. README.md +33 -28
README.md CHANGED
@@ -34,30 +34,35 @@ More details on model performance across various devices, can be found
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
- | AOT-GAN | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 609.651 ms | 0 - 129 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
38
- | AOT-GAN | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 538.394 ms | 1 - 11 MB | NPU | Use Export Script |
39
- | AOT-GAN | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 192.71 ms | 3 - 145 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
40
- | AOT-GAN | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 203.074 ms | 0 - 139 MB | NPU | Use Export Script |
41
- | AOT-GAN | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 146.93 ms | 3 - 28 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
42
- | AOT-GAN | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 118.577 ms | 4 - 6 MB | NPU | Use Export Script |
43
- | AOT-GAN | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 609.651 ms | 0 - 129 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
44
- | AOT-GAN | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 538.394 ms | 1 - 11 MB | NPU | Use Export Script |
45
- | AOT-GAN | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 149.058 ms | 2 - 27 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
46
- | AOT-GAN | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 116.58 ms | 4 - 6 MB | NPU | Use Export Script |
47
- | AOT-GAN | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 217.202 ms | 3 - 121 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
48
- | AOT-GAN | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 175.64 ms | 1 - 19 MB | NPU | Use Export Script |
49
- | AOT-GAN | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 149.171 ms | 3 - 28 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
50
- | AOT-GAN | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 116.698 ms | 4 - 6 MB | NPU | Use Export Script |
51
- | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 146.308 ms | 3 - 31 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
52
- | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 116.306 ms | 4 - 44 MB | NPU | Use Export Script |
53
- | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 115.195 ms | 6 - 108 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
54
- | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 109.318 ms | 2 - 154 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
55
- | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 84.569 ms | 4 - 180 MB | NPU | Use Export Script |
56
- | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 84.506 ms | 0 - 176 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
57
- | AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 109.794 ms | 3 - 134 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
58
- | AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 76.615 ms | 4 - 247 MB | NPU | Use Export Script |
59
- | AOT-GAN | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 120.59 ms | 4 - 4 MB | NPU | Use Export Script |
60
- | AOT-GAN | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 135.973 ms | 32 - 32 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
 
 
 
 
 
61
 
62
 
63
 
@@ -121,8 +126,8 @@ Profiling Results
121
  AOT-GAN
122
  Device : cs_8275 (ANDROID 14)
123
  Runtime : TFLITE
124
- Estimated inference time (ms) : 609.7
125
- Estimated peak memory usage (MB): [0, 129]
126
  Total # Ops : 235
127
  Compute Unit(s) : npu (235 ops) gpu (0 ops) cpu (0 ops)
128
  ```
@@ -211,13 +216,13 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
211
  You can also run the demo on-device.
212
 
213
  ```bash
214
- python -m qai_hub_models.models.aotgan.demo --on-device
215
  ```
216
 
217
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
218
  environment, please add the following to your cell (instead of the above).
219
  ```
220
- %run -m qai_hub_models.models.aotgan.demo -- --on-device
221
  ```
222
 
223
 
 
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
+ | AOT-GAN | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 3728.628 ms | 0 - 135 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
38
+ | AOT-GAN | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 537.852 ms | 2 - 11 MB | NPU | Use Export Script |
39
+ | AOT-GAN | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 191.613 ms | 3 - 152 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
40
+ | AOT-GAN | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 224.689 ms | 4 - 161 MB | NPU | Use Export Script |
41
+ | AOT-GAN | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 149.487 ms | 1 - 25 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
42
+ | AOT-GAN | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 116.628 ms | 4 - 6 MB | NPU | Use Export Script |
43
+ | AOT-GAN | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 196.935 ms | 3 - 138 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
44
+ | AOT-GAN | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 160.101 ms | 1 - 11 MB | NPU | Use Export Script |
45
+ | AOT-GAN | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 3728.628 ms | 0 - 135 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
46
+ | AOT-GAN | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 537.852 ms | 2 - 11 MB | NPU | Use Export Script |
47
+ | AOT-GAN | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 146.129 ms | 3 - 25 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
48
+ | AOT-GAN | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 117.645 ms | 4 - 6 MB | NPU | Use Export Script |
49
+ | AOT-GAN | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 218.507 ms | 3 - 127 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
50
+ | AOT-GAN | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 177.893 ms | 1 - 19 MB | NPU | Use Export Script |
51
+ | AOT-GAN | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 146.754 ms | 3 - 21 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
52
+ | AOT-GAN | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 116.826 ms | 4 - 6 MB | NPU | Use Export Script |
53
+ | AOT-GAN | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 196.935 ms | 3 - 138 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
54
+ | AOT-GAN | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 160.101 ms | 1 - 11 MB | NPU | Use Export Script |
55
+ | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 147.218 ms | 3 - 25 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
56
+ | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 117.168 ms | 4 - 40 MB | NPU | Use Export Script |
57
+ | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 117.81 ms | 6 - 100 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
58
+ | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 110.04 ms | 1 - 158 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
59
+ | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 85.323 ms | 4 - 207 MB | NPU | Use Export Script |
60
+ | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 85.118 ms | 0 - 196 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
61
+ | AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 111.375 ms | 2 - 138 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
62
+ | AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 80.251 ms | 4 - 274 MB | NPU | Use Export Script |
63
+ | AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 67.682 ms | 11 - 280 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
64
+ | AOT-GAN | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 120.563 ms | 4 - 4 MB | NPU | Use Export Script |
65
+ | AOT-GAN | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 136.355 ms | 32 - 32 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
66
 
67
 
68
 
 
126
  AOT-GAN
127
  Device : cs_8275 (ANDROID 14)
128
  Runtime : TFLITE
129
+ Estimated inference time (ms) : 3728.6
130
+ Estimated peak memory usage (MB): [0, 135]
131
  Total # Ops : 235
132
  Compute Unit(s) : npu (235 ops) gpu (0 ops) cpu (0 ops)
133
  ```
 
216
  You can also run the demo on-device.
217
 
218
  ```bash
219
+ python -m qai_hub_models.models.aotgan.demo --eval-mode on-device
220
  ```
221
 
222
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
223
  environment, please add the following to your cell (instead of the above).
224
  ```
225
+ %run -m qai_hub_models.models.aotgan.demo -- --eval-mode on-device
226
  ```
227
 
228