Hugging Face's logo

qualcomm
/
Unet-Segmentation

Image Segmentation
PyTorch
LiteRT
backbone
real_time
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
92df43a
·
verified ·
1 Parent(s): d187f06

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +38 -38
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  library_name: pytorch
3
- license: gpl-3.0
4
  tags:
5
  - backbone
6
  - real_time
@@ -27,7 +27,7 @@ More details on model performance across various devices, can be found
27
 
28
  ### Model Details
29
 
30
- - **Model Type:** Semantic segmentation
31
  - **Model Stats:**
32
  - Model checkpoint: unet_carvana_scale1.0_epoch2
33
  - Input resolution: 224x224
@@ -35,37 +35,37 @@ More details on model performance across various devices, can be found
35
  - Model size: 118 MB
36
  - Number of output classes: 2 (foreground / background)
37
 
38
- | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
39
  |---|---|---|---|---|---|---|---|---|
40
- | Unet-Segmentation | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 164.204 ms | 6 - 462 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
41
- | Unet-Segmentation | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 157.84 ms | 9 - 31 MB | FP16 | NPU | [Unet-Segmentation.so](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.so) |
42
- | Unet-Segmentation | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 169.543 ms | 0 - 145 MB | FP16 | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
43
- | Unet-Segmentation | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 120.926 ms | 7 - 135 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
44
- | Unet-Segmentation | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 120.543 ms | 9 - 89 MB | FP16 | NPU | [Unet-Segmentation.so](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.so) |
45
- | Unet-Segmentation | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 120.86 ms | 21 - 112 MB | FP16 | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
46
- | Unet-Segmentation | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 98.424 ms | 6 - 93 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
47
- | Unet-Segmentation | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 98.122 ms | 9 - 94 MB | FP16 | NPU | Use Export Script |
48
- | Unet-Segmentation | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 111.187 ms | 21 - 112 MB | FP16 | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
49
- | Unet-Segmentation | SA7255P ADP | SA7255P | TFLITE | 7318.406 ms | 0 - 85 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
50
- | Unet-Segmentation | SA7255P ADP | SA7255P | QNN | 7309.112 ms | 4 - 13 MB | FP16 | NPU | Use Export Script |
51
- | Unet-Segmentation | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 161.848 ms | 0 - 485 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
52
- | Unet-Segmentation | SA8255 (Proxy) | SA8255P Proxy | QNN | 156.401 ms | 10 - 12 MB | FP16 | NPU | Use Export Script |
53
- | Unet-Segmentation | SA8295P ADP | SA8295P | TFLITE | 286.65 ms | 6 - 93 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
54
- | Unet-Segmentation | SA8295P ADP | SA8295P | QNN | 277.494 ms | 0 - 17 MB | FP16 | NPU | Use Export Script |
55
- | Unet-Segmentation | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 162.366 ms | 6 - 463 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
56
- | Unet-Segmentation | SA8650 (Proxy) | SA8650P Proxy | QNN | 154.934 ms | 10 - 12 MB | FP16 | NPU | Use Export Script |
57
- | Unet-Segmentation | SA8775P ADP | SA8775P | TFLITE | 313.094 ms | 6 - 91 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
58
- | Unet-Segmentation | SA8775P ADP | SA8775P | QNN | 308.285 ms | 2 - 11 MB | FP16 | NPU | Use Export Script |
59
- | Unet-Segmentation | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 7318.406 ms | 0 - 85 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
60
- | Unet-Segmentation | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 7309.112 ms | 4 - 13 MB | FP16 | NPU | Use Export Script |
61
- | Unet-Segmentation | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 163.481 ms | 6 - 462 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
62
- | Unet-Segmentation | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 155.071 ms | 10 - 19 MB | FP16 | NPU | Use Export Script |
63
- | Unet-Segmentation | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 313.094 ms | 6 - 91 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
64
- | Unet-Segmentation | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 308.285 ms | 2 - 11 MB | FP16 | NPU | Use Export Script |
65
- | Unet-Segmentation | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 304.614 ms | 6 - 154 MB | FP16 | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
66
- | Unet-Segmentation | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 308.237 ms | 8 - 108 MB | FP16 | NPU | Use Export Script |
67
- | Unet-Segmentation | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 145.554 ms | 9 - 9 MB | FP16 | NPU | Use Export Script |
68
- | Unet-Segmentation | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 153.173 ms | 53 - 53 MB | FP16 | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
69
 
70
 
71
 
@@ -127,12 +127,12 @@ python -m qai_hub_models.models.unet_segmentation.export
127
  Profiling Results
128
  ------------------------------------------------------------
129
  Unet-Segmentation
130
- Device : Samsung Galaxy S23 (13)
131
- Runtime : TFLITE
132
- Estimated inference time (ms) : 164.2
133
- Estimated peak memory usage (MB): [6, 462]
134
- Total # Ops : 32
135
- Compute Unit(s) : NPU (32 ops)
136
  ```
137
 
138
 
 
1
  ---
2
  library_name: pytorch
3
+ license: other
4
  tags:
5
  - backbone
6
  - real_time
 
27
 
28
  ### Model Details
29
 
30
+ - **Model Type:** Model_use_case.semantic_segmentation
31
  - **Model Stats:**
32
  - Model checkpoint: unet_carvana_scale1.0_epoch2
33
  - Input resolution: 224x224
 
35
  - Model size: 118 MB
36
  - Number of output classes: 2 (foreground / background)
37
 
38
+ | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
39
  |---|---|---|---|---|---|---|---|---|
40
+ | Unet-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 959.769 ms | 0 - 85 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
41
+ | Unet-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 953.53 ms | 4 - 14 MB | NPU | Use Export Script |
42
+ | Unet-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 294.466 ms | 6 - 153 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
43
+ | Unet-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 301.944 ms | 9 - 117 MB | NPU | Use Export Script |
44
+ | Unet-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 161.723 ms | 6 - 463 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
45
+ | Unet-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 160.301 ms | 10 - 13 MB | NPU | Use Export Script |
46
+ | Unet-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 258.565 ms | 6 - 91 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
47
+ | Unet-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 253.879 ms | 1 - 13 MB | NPU | Use Export Script |
48
+ | Unet-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 959.769 ms | 0 - 85 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
49
+ | Unet-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 953.53 ms | 4 - 14 MB | NPU | Use Export Script |
50
+ | Unet-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 161.059 ms | 6 - 462 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
51
+ | Unet-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 153.977 ms | 10 - 12 MB | NPU | Use Export Script |
52
+ | Unet-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 286.646 ms | 6 - 93 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
53
+ | Unet-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 277.483 ms | 0 - 18 MB | NPU | Use Export Script |
54
+ | Unet-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 163.936 ms | 6 - 464 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
55
+ | Unet-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 152.948 ms | 11 - 14 MB | NPU | Use Export Script |
56
+ | Unet-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 258.565 ms | 6 - 91 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
57
+ | Unet-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 253.879 ms | 1 - 13 MB | NPU | Use Export Script |
58
+ | Unet-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 162.184 ms | 6 - 462 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
59
+ | Unet-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 155.536 ms | 8 - 45 MB | NPU | Use Export Script |
60
+ | Unet-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 166.412 ms | 13 - 221 MB | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
61
+ | Unet-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 120.562 ms | 4 - 130 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
62
+ | Unet-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 116.097 ms | 9 - 110 MB | NPU | Use Export Script |
63
+ | Unet-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 123.289 ms | 22 - 114 MB | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
64
+ | Unet-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 102.884 ms | 5 - 91 MB | NPU | [Unet-Segmentation.tflite](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.tflite) |
65
+ | Unet-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 107.411 ms | 9 - 109 MB | NPU | Use Export Script |
66
+ | Unet-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 110.866 ms | 21 - 112 MB | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
67
+ | Unet-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 145.676 ms | 9 - 9 MB | NPU | Use Export Script |
68
+ | Unet-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 153.193 ms | 54 - 54 MB | NPU | [Unet-Segmentation.onnx](https://huggingface.co/qualcomm/Unet-Segmentation/blob/main/Unet-Segmentation.onnx) |
69
 
70
 
71
 
 
127
  Profiling Results
128
  ------------------------------------------------------------
129
  Unet-Segmentation
130
+ Device : cs_8275 (ANDROID 14)
131
+ Runtime : TFLITE
132
+ Estimated inference time (ms) : 959.8
133
+ Estimated peak memory usage (MB): [0, 85]
134
+ Total # Ops : 32
135
+ Compute Unit(s) : npu (32 ops) gpu (0 ops) cpu (0 ops)
136
  ```
137
 
138