Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -35,45 +35,44 @@ 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 |
-
| Mobile_Vit | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 5.
|
39 |
-
| Mobile_Vit | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 5.
|
40 |
-
| Mobile_Vit | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 6.
|
41 |
-
| Mobile_Vit | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 3.
|
42 |
-
| Mobile_Vit | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 3.
|
43 |
-
| Mobile_Vit | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 4.
|
44 |
-
| Mobile_Vit | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE |
|
45 |
-
| Mobile_Vit | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 3.
|
46 |
-
| Mobile_Vit | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX |
|
47 |
-
| Mobile_Vit | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 5.
|
48 |
-
| Mobile_Vit | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 4.
|
49 |
-
| Mobile_Vit | SA7255P ADP | SA7255P | TFLITE | 43.
|
50 |
-
| Mobile_Vit | SA7255P ADP | SA7255P | QNN | 43.
|
51 |
-
| Mobile_Vit | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 5.
|
52 |
-
| Mobile_Vit | SA8255 (Proxy) | SA8255P Proxy | QNN | 4.
|
53 |
-
| Mobile_Vit | SA8295P ADP | SA8295P | TFLITE | 8.
|
54 |
-
| Mobile_Vit | SA8295P ADP | SA8295P | QNN | 8.
|
55 |
-
| Mobile_Vit | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 5.
|
56 |
-
| Mobile_Vit | SA8650 (Proxy) | SA8650P Proxy | QNN | 4.
|
57 |
-
| Mobile_Vit | SA8775P ADP | SA8775P | TFLITE | 7.
|
58 |
-
| Mobile_Vit | SA8775P ADP | SA8775P | QNN | 7.
|
59 |
-
| Mobile_Vit | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 7.
|
60 |
-
| Mobile_Vit | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 7.
|
61 |
-
| Mobile_Vit | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 5.
|
62 |
-
| Mobile_Vit | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.
|
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 |
```
|
74 |
|
75 |
|
76 |
-
|
77 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
78 |
|
79 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
@@ -152,7 +151,7 @@ from qai_hub_models.models.mobile_vit import Model
|
|
152 |
torch_model = Model.from_pretrained()
|
153 |
|
154 |
# Device
|
155 |
-
device = hub.Device("Samsung Galaxy
|
156 |
|
157 |
# Trace model
|
158 |
input_shape = torch_model.get_input_spec()
|
@@ -244,7 +243,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
|
|
244 |
|
245 |
|
246 |
## License
|
247 |
-
* The license for the original implementation of Mobile_Vit can be found
|
|
|
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 |
|
|
|
35 |
|
36 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
37 |
|---|---|---|---|---|---|---|---|---|
|
38 |
+
| Mobile_Vit | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 5.178 ms | 0 - 16 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
39 |
+
| Mobile_Vit | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 5.097 ms | 0 - 20 MB | FP16 | NPU | [Mobile_Vit.so](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.so) |
|
40 |
+
| Mobile_Vit | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 6.252 ms | 0 - 47 MB | FP16 | NPU | [Mobile_Vit.onnx](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.onnx) |
|
41 |
+
| Mobile_Vit | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 3.585 ms | 0 - 35 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
42 |
+
| Mobile_Vit | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 3.516 ms | 0 - 38 MB | FP16 | NPU | [Mobile_Vit.so](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.so) |
|
43 |
+
| Mobile_Vit | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 4.351 ms | 1 - 50 MB | FP16 | NPU | [Mobile_Vit.onnx](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.onnx) |
|
44 |
+
| Mobile_Vit | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 3.55 ms | 0 - 36 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
45 |
+
| Mobile_Vit | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 3.483 ms | 1 - 41 MB | FP16 | NPU | Use Export Script |
|
46 |
+
| Mobile_Vit | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 4.05 ms | 0 - 44 MB | FP16 | NPU | [Mobile_Vit.onnx](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.onnx) |
|
47 |
+
| Mobile_Vit | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 5.191 ms | 0 - 16 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
48 |
+
| Mobile_Vit | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 4.637 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
|
49 |
+
| Mobile_Vit | SA7255P ADP | SA7255P | TFLITE | 43.187 ms | 0 - 29 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
50 |
+
| Mobile_Vit | SA7255P ADP | SA7255P | QNN | 43.432 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
|
51 |
+
| Mobile_Vit | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 5.201 ms | 0 - 15 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
52 |
+
| Mobile_Vit | SA8255 (Proxy) | SA8255P Proxy | QNN | 4.657 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
|
53 |
+
| Mobile_Vit | SA8295P ADP | SA8295P | TFLITE | 8.334 ms | 0 - 34 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
54 |
+
| Mobile_Vit | SA8295P ADP | SA8295P | QNN | 8.364 ms | 1 - 15 MB | FP16 | NPU | Use Export Script |
|
55 |
+
| Mobile_Vit | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 5.19 ms | 0 - 15 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
56 |
+
| Mobile_Vit | SA8650 (Proxy) | SA8650P Proxy | QNN | 4.646 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
|
57 |
+
| Mobile_Vit | SA8775P ADP | SA8775P | TFLITE | 7.416 ms | 0 - 29 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
58 |
+
| Mobile_Vit | SA8775P ADP | SA8775P | QNN | 7.708 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
|
59 |
+
| Mobile_Vit | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 7.175 ms | 0 - 34 MB | FP16 | NPU | [Mobile_Vit.tflite](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.tflite) |
|
60 |
+
| Mobile_Vit | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 7.033 ms | 1 - 38 MB | FP16 | NPU | Use Export Script |
|
61 |
+
| Mobile_Vit | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 5.025 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
|
62 |
+
| Mobile_Vit | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.5 ms | 15 - 15 MB | FP16 | NPU | [Mobile_Vit.onnx](https://huggingface.co/qualcomm/Mobile_Vit/blob/main/Mobile_Vit.onnx) |
|
63 |
|
64 |
|
65 |
|
66 |
|
67 |
## Installation
|
68 |
|
|
|
69 |
|
70 |
+
Install the package via pip:
|
71 |
```bash
|
72 |
+
pip install "qai-hub-models[mobile-vit]"
|
73 |
```
|
74 |
|
75 |
|
|
|
76 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
77 |
|
78 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
|
|
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 Mobile_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 |
|