Hugging Face's logo

qualcomm
/
VIT

Image Classification
PyTorch
TF Lite
ONNX
backbone
android
Model card Files Community
qaihm-bot commited on
Commit
36c6a95
·
verified ·
1 Parent(s): 0c54e25

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +31 -30
README.md CHANGED
@@ -35,39 +35,39 @@ More details on model performance across various devices, can be found
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
- | VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 17.011 ms | 0 - 44 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
39
- | VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 17.987 ms | 0 - 41 MB | FP16 | NPU | [VIT.so](https://huggingface.co/qualcomm/VIT/blob/main/VIT.so) |
40
- | VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 14.739 ms | 0 - 361 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
41
- | VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 12.031 ms | 0 - 62 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
42
- | VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 12.462 ms | 1 - 63 MB | FP16 | NPU | [VIT.so](https://huggingface.co/qualcomm/VIT/blob/main/VIT.so) |
43
- | VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 10.636 ms | 1 - 46 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
44
- | VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 11.371 ms | 0 - 62 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
45
- | VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 11.63 ms | 0 - 60 MB | FP16 | NPU | Use Export Script |
46
- | VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 8.623 ms | 0 - 42 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
47
- | VIT | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 17.36 ms | 0 - 46 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
48
- | VIT | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 15.804 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
49
- | VIT | SA7255P ADP | SA7255P | TFLITE | 258.767 ms | 0 - 58 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
50
- | VIT | SA7255P ADP | SA7255P | QNN | 252.384 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
51
- | VIT | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 17.328 ms | 0 - 42 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
52
- | VIT | SA8255 (Proxy) | SA8255P Proxy | QNN | 15.76 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
53
- | VIT | SA8295P ADP | SA8295P | TFLITE | 27.62 ms | 0 - 57 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
54
- | VIT | SA8295P ADP | SA8295P | QNN | 22.953 ms | 1 - 15 MB | FP16 | NPU | Use Export Script |
55
- | VIT | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 17.072 ms | 0 - 47 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
56
- | VIT | SA8650 (Proxy) | SA8650P Proxy | QNN | 15.821 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
57
- | VIT | SA8775P ADP | SA8775P | TFLITE | 24.673 ms | 0 - 58 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
58
- | VIT | SA8775P ADP | SA8775P | QNN | 22.801 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
59
- | VIT | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 24.446 ms | 0 - 58 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
60
- | VIT | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 23.556 ms | 1 - 57 MB | FP16 | NPU | Use Export Script |
61
- | VIT | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 16.18 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
62
- | VIT | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 20.493 ms | 172 - 172 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
63
 
64
 
65
 
66
 
67
  ## Installation
68
 
69
- This model can be installed as a Python package via pip.
70
 
 
71
  ```bash
72
  pip install qai-hub-models
73
  ```
@@ -123,8 +123,8 @@ Profiling Results
123
  VIT
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
- Estimated inference time (ms) : 17.0
127
- Estimated peak memory usage (MB): [0, 44]
128
  Total # Ops : 1579
129
  Compute Unit(s) : NPU (1579 ops)
130
  ```
@@ -151,7 +151,7 @@ from qai_hub_models.models.vit import Model
151
  torch_model = Model.from_pretrained()
152
 
153
  # Device
154
- device = hub.Device("Samsung Galaxy S23")
155
 
156
  # Trace model
157
  input_shape = torch_model.get_input_spec()
@@ -243,7 +243,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
243
 
244
 
245
  ## License
246
- * The license for the original implementation of VIT can be found [here](https://github.com/pytorch/vision/blob/main/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
 
 
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 17.377 ms | 0 - 46 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
39
+ | VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 17.91 ms | 1 - 48 MB | FP16 | NPU | [VIT.so](https://huggingface.co/qualcomm/VIT/blob/main/VIT.so) |
40
+ | VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 14.754 ms | 0 - 361 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
41
+ | VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 12.044 ms | 0 - 62 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
42
+ | VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 12.434 ms | 1 - 57 MB | FP16 | NPU | [VIT.so](https://huggingface.co/qualcomm/VIT/blob/main/VIT.so) |
43
+ | VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 10.645 ms | 0 - 46 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
44
+ | VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 11.399 ms | 0 - 63 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
45
+ | VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 9.734 ms | 1 - 62 MB | FP16 | NPU | Use Export Script |
46
+ | VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 8.697 ms | 0 - 43 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
47
+ | VIT | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 17.427 ms | 0 - 45 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
48
+ | VIT | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 15.73 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
49
+ | VIT | SA7255P ADP | SA7255P | TFLITE | 258.686 ms | 0 - 58 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
50
+ | VIT | SA7255P ADP | SA7255P | QNN | 252.191 ms | 1 - 9 MB | FP16 | NPU | Use Export Script |
51
+ | VIT | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 17.313 ms | 0 - 45 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
52
+ | VIT | SA8255 (Proxy) | SA8255P Proxy | QNN | 15.795 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
53
+ | VIT | SA8295P ADP | SA8295P | TFLITE | 27.626 ms | 0 - 57 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
54
+ | VIT | SA8295P ADP | SA8295P | QNN | 22.986 ms | 1 - 15 MB | FP16 | NPU | Use Export Script |
55
+ | VIT | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 17.339 ms | 0 - 46 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
56
+ | VIT | SA8650 (Proxy) | SA8650P Proxy | QNN | 15.808 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
57
+ | VIT | SA8775P ADP | SA8775P | TFLITE | 24.668 ms | 0 - 58 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
58
+ | VIT | SA8775P ADP | SA8775P | QNN | 22.815 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
59
+ | VIT | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 24.482 ms | 0 - 59 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
60
+ | VIT | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 23.502 ms | 1 - 54 MB | FP16 | NPU | Use Export Script |
61
+ | VIT | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 16.177 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
62
+ | VIT | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 20.528 ms | 171 - 171 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
63
 
64
 
65
 
66
 
67
  ## Installation
68
 
 
69
 
70
+ Install the package via pip:
71
  ```bash
72
  pip install qai-hub-models
73
  ```
 
123
  VIT
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
+ Estimated inference time (ms) : 17.4
127
+ Estimated peak memory usage (MB): [0, 46]
128
  Total # Ops : 1579
129
  Compute Unit(s) : NPU (1579 ops)
130
  ```
 
151
  torch_model = Model.from_pretrained()
152
 
153
  # Device
154
+ device = hub.Device("Samsung Galaxy S24")
155
 
156
  # Trace model
157
  input_shape = torch_model.get_input_spec()
 
243
 
244
 
245
  ## License
246
+ * The license for the original implementation of VIT can be found
247
+ [here](https://github.com/pytorch/vision/blob/main/LICENSE).
248
  * 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)
249
 
250