Hugging Face's logo

qualcomm
/
Swin-Tiny

Image Classification
PyTorch
LiteRT
backbone
bu_auto
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
570621d
·
verified ·
1 Parent(s): cfad495

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +43 -38
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  library_name: pytorch
3
- license: bsd-3-clause
4
  tags:
5
  - backbone
6
  - android
@@ -26,43 +26,48 @@ More details on model performance across various devices, can be found
26
 
27
  ### Model Details
28
 
29
- - **Model Type:** Image classification
30
  - **Model Stats:**
31
  - Model checkpoint: Imagenet
32
  - Input resolution: 224x224
33
  - Number of parameters: 28.8M
34
- - Model size: 110 MB
35
 
36
- | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
- | Swin-Tiny | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 11.068 ms | 0 - 15 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
39
- | Swin-Tiny | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 11.531 ms | 0 - 15 MB | FP16 | NPU | [Swin-Tiny.so](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.so) |
40
- | Swin-Tiny | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 10.702 ms | 0 - 199 MB | FP16 | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.onnx) |
41
- | Swin-Tiny | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 7.409 ms | 0 - 161 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
42
- | Swin-Tiny | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 7.996 ms | 0 - 157 MB | FP16 | NPU | [Swin-Tiny.so](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.so) |
43
- | Swin-Tiny | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 7.4 ms | 1 - 168 MB | FP16 | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.onnx) |
44
- | Swin-Tiny | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 6.804 ms | 0 - 154 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
45
- | Swin-Tiny | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 7.157 ms | 0 - 150 MB | FP16 | NPU | Use Export Script |
46
- | Swin-Tiny | SA7255P ADP | SA7255P | TFLITE | 93.625 ms | 0 - 154 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
47
- | Swin-Tiny | SA7255P ADP | SA7255P | QNN | 94.56 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
48
- | Swin-Tiny | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 11.037 ms | 0 - 15 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
49
- | Swin-Tiny | SA8255 (Proxy) | SA8255P Proxy | QNN | 11.003 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
50
- | Swin-Tiny | SA8295P ADP | SA8295P | TFLITE | 15.25 ms | 0 - 150 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
51
- | Swin-Tiny | SA8295P ADP | SA8295P | QNN | 15.138 ms | 1 - 19 MB | FP16 | NPU | Use Export Script |
52
- | Swin-Tiny | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 11.062 ms | 0 - 15 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
53
- | Swin-Tiny | SA8650 (Proxy) | SA8650P Proxy | QNN | 11.043 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
54
- | Swin-Tiny | SA8775P ADP | SA8775P | TFLITE | 14.941 ms | 0 - 153 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
55
- | Swin-Tiny | SA8775P ADP | SA8775P | QNN | 14.648 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
56
- | Swin-Tiny | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 93.625 ms | 0 - 154 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
57
- | Swin-Tiny | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 94.56 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
58
- | Swin-Tiny | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 11.029 ms | 0 - 15 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
59
- | Swin-Tiny | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 11.006 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
60
- | Swin-Tiny | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 14.941 ms | 0 - 153 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
61
- | Swin-Tiny | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 14.648 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
62
- | Swin-Tiny | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 13.812 ms | 0 - 159 MB | FP16 | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
63
- | Swin-Tiny | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 14.43 ms | 1 - 155 MB | FP16 | NPU | Use Export Script |
64
- | Swin-Tiny | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 11.437 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
65
- | Swin-Tiny | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 11.737 ms | 57 - 57 MB | FP16 | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.onnx) |
 
 
 
 
 
66
 
67
 
68
 
@@ -124,12 +129,12 @@ python -m qai_hub_models.models.swin_tiny.export
124
  Profiling Results
125
  ------------------------------------------------------------
126
  Swin-Tiny
127
- Device : Samsung Galaxy S23 (13)
128
- Runtime : TFLITE
129
- Estimated inference time (ms) : 11.1
130
- Estimated peak memory usage (MB): [0, 15]
131
- Total # Ops : 837
132
- Compute Unit(s) : NPU (837 ops)
133
  ```
134
 
135
 
 
1
  ---
2
  library_name: pytorch
3
+ license: other
4
  tags:
5
  - backbone
6
  - android
 
26
 
27
  ### Model Details
28
 
29
+ - **Model Type:** Model_use_case.image_classification
30
  - **Model Stats:**
31
  - Model checkpoint: Imagenet
32
  - Input resolution: 224x224
33
  - Number of parameters: 28.8M
34
+ - Model size (float): 110 MB
35
 
36
+ | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | Swin-Tiny | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 24.985 ms | 0 - 152 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
39
+ | Swin-Tiny | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 24.412 ms | 1 - 10 MB | NPU | Use Export Script |
40
+ | Swin-Tiny | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 13.901 ms | 0 - 160 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
41
+ | Swin-Tiny | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 16.41 ms | 0 - 158 MB | NPU | Use Export Script |
42
+ | Swin-Tiny | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 11.03 ms | 0 - 14 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
43
+ | Swin-Tiny | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 10.921 ms | 1 - 4 MB | NPU | Use Export Script |
44
+ | Swin-Tiny | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 12.422 ms | 0 - 153 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
45
+ | Swin-Tiny | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 12.182 ms | 1 - 15 MB | NPU | Use Export Script |
46
+ | Swin-Tiny | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 24.985 ms | 0 - 152 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
47
+ | Swin-Tiny | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 24.412 ms | 1 - 10 MB | NPU | Use Export Script |
48
+ | Swin-Tiny | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 11.083 ms | 0 - 15 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
49
+ | Swin-Tiny | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 11.011 ms | 1 - 3 MB | NPU | Use Export Script |
50
+ | Swin-Tiny | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 15.284 ms | 0 - 151 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
51
+ | Swin-Tiny | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 15.108 ms | 1 - 18 MB | NPU | Use Export Script |
52
+ | Swin-Tiny | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 11.05 ms | 0 - 15 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
53
+ | Swin-Tiny | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 11.0 ms | 1 - 3 MB | NPU | Use Export Script |
54
+ | Swin-Tiny | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 12.422 ms | 0 - 153 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
55
+ | Swin-Tiny | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 12.182 ms | 1 - 15 MB | NPU | Use Export Script |
56
+ | Swin-Tiny | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 11.026 ms | 0 - 14 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
57
+ | Swin-Tiny | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 10.916 ms | 0 - 40 MB | NPU | Use Export Script |
58
+ | Swin-Tiny | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 9.696 ms | 0 - 149 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.onnx) |
59
+ | Swin-Tiny | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 7.363 ms | 0 - 160 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
60
+ | Swin-Tiny | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 7.371 ms | 1 - 158 MB | NPU | Use Export Script |
61
+ | Swin-Tiny | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 6.396 ms | 1 - 156 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.onnx) |
62
+ | Swin-Tiny | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 6.8 ms | 0 - 154 MB | NPU | [Swin-Tiny.tflite](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.tflite) |
63
+ | Swin-Tiny | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 6.75 ms | 1 - 152 MB | NPU | Use Export Script |
64
+ | Swin-Tiny | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 6.093 ms | 0 - 146 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.onnx) |
65
+ | Swin-Tiny | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 11.44 ms | 1 - 1 MB | NPU | Use Export Script |
66
+ | Swin-Tiny | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 11.385 ms | 57 - 57 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny.onnx) |
67
+ | Swin-Tiny | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 59.081 ms | 152 - 275 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny_w8a16.onnx) |
68
+ | Swin-Tiny | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 55.801 ms | 160 - 323 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny_w8a16.onnx) |
69
+ | Swin-Tiny | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 48.089 ms | 158 - 290 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny_w8a16.onnx) |
70
+ | Swin-Tiny | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 67.552 ms | 223 - 223 MB | NPU | [Swin-Tiny.onnx](https://huggingface.co/qualcomm/Swin-Tiny/blob/main/Swin-Tiny_w8a16.onnx) |
71
 
72
 
73
 
 
129
  Profiling Results
130
  ------------------------------------------------------------
131
  Swin-Tiny
132
+ Device : cs_8275 (ANDROID 14)
133
+ Runtime : TFLITE
134
+ Estimated inference time (ms) : 25.0
135
+ Estimated peak memory usage (MB): [0, 152]
136
+ Total # Ops : 837
137
+ Compute Unit(s) : npu (837 ops) gpu (0 ops) cpu (0 ops)
138
  ```
139
 
140