Hugging Face's logo

qualcomm
/
SINet

Image Segmentation
PyTorch
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
41a48d4
·
verified ·
1 Parent(s): 7863346

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: mit
4
  tags:
5
  - android
6
  pipeline_tag: image-segmentation
@@ -25,7 +25,7 @@ More details on model performance across various devices, can be found
25
 
26
  ### Model Details
27
 
28
- - **Model Type:** Semantic segmentation
29
  - **Model Stats:**
30
  - Model checkpoint: SINet.pth
31
  - Input resolution: 224x224
@@ -33,37 +33,37 @@ More details on model performance across various devices, can be found
33
  - Model size: 430 KB
34
  - Number of output classes: 2 (foreground / background)
35
 
36
- | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
- | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.706 ms | 0 - 7 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
39
- | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 1.702 ms | 1 - 7 MB | FP16 | NPU | [SINet.so](https://huggingface.co/qualcomm/SINet/blob/main/SINet.so) |
40
- | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.292 ms | 0 - 9 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
41
- | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.148 ms | 0 - 23 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
42
- | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.131 ms | 0 - 23 MB | FP16 | NPU | [SINet.so](https://huggingface.co/qualcomm/SINet/blob/main/SINet.so) |
43
- | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.521 ms | 0 - 24 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
44
- | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.87 ms | 0 - 22 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
45
- | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.032 ms | 1 - 20 MB | FP16 | NPU | Use Export Script |
46
- | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.515 ms | 0 - 19 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
47
- | SINet | SA7255P ADP | SA7255P | TFLITE | 9.871 ms | 0 - 15 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
48
- | SINet | SA7255P ADP | SA7255P | QNN | 9.521 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
49
- | SINet | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.714 ms | 0 - 6 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
50
- | SINet | SA8255 (Proxy) | SA8255P Proxy | QNN | 1.647 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
51
- | SINet | SA8295P ADP | SA8295P | TFLITE | 2.411 ms | 0 - 19 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
52
- | SINet | SA8295P ADP | SA8295P | QNN | 2.388 ms | 0 - 17 MB | FP16 | NPU | Use Export Script |
53
- | SINet | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.716 ms | 0 - 7 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
54
- | SINet | SA8650 (Proxy) | SA8650P Proxy | QNN | 1.655 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
55
- | SINet | SA8775P ADP | SA8775P | TFLITE | 2.631 ms | 0 - 15 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
56
- | SINet | SA8775P ADP | SA8775P | QNN | 2.517 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
57
- | SINet | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 9.871 ms | 0 - 15 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
58
- | SINet | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 9.521 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
59
- | SINet | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.717 ms | 0 - 7 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
60
- | SINet | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 1.616 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
61
- | SINet | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 2.631 ms | 0 - 15 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
62
- | SINet | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 2.517 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
63
- | SINet | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.974 ms | 0 - 29 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
64
- | SINet | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 1.938 ms | 1 - 28 MB | FP16 | NPU | Use Export Script |
65
- | SINet | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.896 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
66
- | SINet | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.412 ms | 2 - 2 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
67
 
68
 
69
 
@@ -125,12 +125,12 @@ python -m qai_hub_models.models.sinet.export
125
  Profiling Results
126
  ------------------------------------------------------------
127
  SINet
128
- Device : Samsung Galaxy S23 (13)
129
- Runtime : TFLITE
130
- Estimated inference time (ms) : 1.7
131
- Estimated peak memory usage (MB): [0, 7]
132
- Total # Ops : 222
133
- Compute Unit(s) : NPU (222 ops)
134
  ```
135
 
136
 
 
1
  ---
2
  library_name: pytorch
3
+ license: other
4
  tags:
5
  - android
6
  pipeline_tag: image-segmentation
 
25
 
26
  ### Model Details
27
 
28
+ - **Model Type:** Model_use_case.semantic_segmentation
29
  - **Model Stats:**
30
  - Model checkpoint: SINet.pth
31
  - Input resolution: 224x224
 
33
  - Model size: 430 KB
34
  - Number of output classes: 2 (foreground / background)
35
 
36
+ | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | SINet | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 3.824 ms | 0 - 15 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
39
+ | SINet | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 3.569 ms | 1 - 10 MB | NPU | Use Export Script |
40
+ | SINet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 1.932 ms | 0 - 21 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
41
+ | SINet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 2.513 ms | 1 - 35 MB | NPU | Use Export Script |
42
+ | SINet | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 1.7 ms | 0 - 7 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
43
+ | SINet | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 1.627 ms | 1 - 4 MB | NPU | Use Export Script |
44
+ | SINet | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 2.21 ms | 0 - 17 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
45
+ | SINet | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 2.129 ms | 1 - 15 MB | NPU | Use Export Script |
46
+ | SINet | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 3.824 ms | 0 - 15 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
47
+ | SINet | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 3.569 ms | 1 - 10 MB | NPU | Use Export Script |
48
+ | SINet | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 1.699 ms | 0 - 6 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
49
+ | SINet | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 1.654 ms | 1 - 4 MB | NPU | Use Export Script |
50
+ | SINet | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 2.38 ms | 0 - 18 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
51
+ | SINet | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 2.37 ms | 0 - 18 MB | NPU | Use Export Script |
52
+ | SINet | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 1.706 ms | 0 - 6 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
53
+ | SINet | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 1.641 ms | 1 - 3 MB | NPU | Use Export Script |
54
+ | SINet | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 2.21 ms | 0 - 17 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
55
+ | SINet | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 2.129 ms | 1 - 15 MB | NPU | Use Export Script |
56
+ | SINet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 1.71 ms | 0 - 6 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
57
+ | SINet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 1.666 ms | 1 - 8 MB | NPU | Use Export Script |
58
+ | SINet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 2.264 ms | 0 - 9 MB | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
59
+ | SINet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1.154 ms | 0 - 31 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
60
+ | SINet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 1.117 ms | 0 - 28 MB | NPU | Use Export Script |
61
+ | SINet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 1.51 ms | 0 - 32 MB | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
62
+ | SINet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 1.054 ms | 0 - 19 MB | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
63
+ | SINet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 1.005 ms | 1 - 23 MB | NPU | Use Export Script |
64
+ | SINet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 1.533 ms | 1 - 20 MB | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
65
+ | SINet | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.909 ms | 1 - 1 MB | NPU | Use Export Script |
66
+ | SINet | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.415 ms | 2 - 2 MB | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
67
 
68
 
69
 
 
125
  Profiling Results
126
  ------------------------------------------------------------
127
  SINet
128
+ Device : cs_8275 (ANDROID 14)
129
+ Runtime : TFLITE
130
+ Estimated inference time (ms) : 3.8
131
+ Estimated peak memory usage (MB): [0, 15]
132
+ Total # Ops : 222
133
+ Compute Unit(s) : npu (222 ops) gpu (0 ops) cpu (0 ops)
134
  ```
135
 
136