Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -34,45 +34,44 @@ More details on model performance across various devices, can be found
|
|
34 |
|
35 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
36 |
|---|---|---|---|---|---|---|---|---|
|
37 |
-
| Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE |
|
38 |
-
| Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN |
|
39 |
-
| Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX |
|
40 |
-
| Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE |
|
41 |
-
| Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN |
|
42 |
-
| Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX |
|
43 |
-
| Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE |
|
44 |
-
| Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN |
|
45 |
-
| Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX |
|
46 |
-
| Depth-Anything | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE |
|
47 |
-
| Depth-Anything | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 229.
|
48 |
-
| Depth-Anything | SA7255P ADP | SA7255P | TFLITE |
|
49 |
-
| Depth-Anything | SA7255P ADP | SA7255P | QNN |
|
50 |
-
| Depth-Anything | SA8255 (Proxy) | SA8255P Proxy | TFLITE |
|
51 |
-
| Depth-Anything | SA8255 (Proxy) | SA8255P Proxy | QNN |
|
52 |
-
| Depth-Anything | SA8295P ADP | SA8295P | TFLITE | 387.
|
53 |
-
| Depth-Anything | SA8295P ADP | SA8295P | QNN |
|
54 |
-
| Depth-Anything | SA8650 (Proxy) | SA8650P Proxy | TFLITE |
|
55 |
-
| Depth-Anything | SA8650 (Proxy) | SA8650P Proxy | QNN |
|
56 |
-
| Depth-Anything | SA8775P ADP | SA8775P | TFLITE | 368.
|
57 |
-
| Depth-Anything | SA8775P ADP | SA8775P | QNN |
|
58 |
-
| Depth-Anything | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE |
|
59 |
-
| Depth-Anything | QCS8450 (Proxy) | QCS8450 Proxy | QNN |
|
60 |
-
| Depth-Anything | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN |
|
61 |
-
| Depth-Anything | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 271.
|
62 |
|
63 |
|
64 |
|
65 |
|
66 |
## Installation
|
67 |
|
68 |
-
This model can be installed as a Python package via pip.
|
69 |
|
|
|
70 |
```bash
|
71 |
-
pip install "qai-hub-models[
|
72 |
```
|
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
|
@@ -123,10 +122,10 @@ Profiling Results
|
|
123 |
Depth-Anything
|
124 |
Device : Samsung Galaxy S23 (13)
|
125 |
Runtime : TFLITE
|
126 |
-
Estimated inference time (ms) :
|
127 |
-
Estimated peak memory usage (MB): [1,
|
128 |
-
Total # Ops :
|
129 |
-
Compute Unit(s) : NPU (
|
130 |
```
|
131 |
|
132 |
|
@@ -151,7 +150,7 @@ from qai_hub_models.models.depth_anything import Model
|
|
151 |
torch_model = Model.from_pretrained()
|
152 |
|
153 |
# Device
|
154 |
-
device = hub.Device("Samsung Galaxy
|
155 |
|
156 |
# Trace model
|
157 |
input_shape = torch_model.get_input_spec()
|
@@ -243,7 +242,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
|
|
243 |
|
244 |
|
245 |
## License
|
246 |
-
* The license for the original implementation of Depth-Anything can be found
|
|
|
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 |
|
|
|
34 |
|
35 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
36 |
|---|---|---|---|---|---|---|---|---|
|
37 |
+
| Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 327.858 ms | 1 - 60 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
38 |
+
| Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 379.27 ms | 3 - 83 MB | FP16 | NPU | [Depth-Anything.so](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.so) |
|
39 |
+
| Depth-Anything | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 240.144 ms | 0 - 181 MB | FP16 | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
|
40 |
+
| Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 250.711 ms | 0 - 246 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
41 |
+
| Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 286.215 ms | 0 - 258 MB | FP16 | NPU | [Depth-Anything.so](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.so) |
|
42 |
+
| Depth-Anything | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 186.098 ms | 5 - 280 MB | FP16 | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
|
43 |
+
| Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 236.545 ms | 1 - 272 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
44 |
+
| Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 242.956 ms | 3 - 283 MB | FP16 | NPU | Use Export Script |
|
45 |
+
| Depth-Anything | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 158.768 ms | 5 - 297 MB | FP16 | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
|
46 |
+
| Depth-Anything | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 317.382 ms | 0 - 58 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
47 |
+
| Depth-Anything | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 229.212 ms | 3 - 6 MB | FP16 | NPU | Use Export Script |
|
48 |
+
| Depth-Anything | SA7255P ADP | SA7255P | TFLITE | 1141.05 ms | 0 - 268 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
49 |
+
| Depth-Anything | SA7255P ADP | SA7255P | QNN | 1005.592 ms | 3 - 12 MB | FP16 | NPU | Use Export Script |
|
50 |
+
| Depth-Anything | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 322.962 ms | 0 - 61 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
51 |
+
| Depth-Anything | SA8255 (Proxy) | SA8255P Proxy | QNN | 230.498 ms | 3 - 5 MB | FP16 | NPU | Use Export Script |
|
52 |
+
| Depth-Anything | SA8295P ADP | SA8295P | TFLITE | 387.756 ms | 1 - 271 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
53 |
+
| Depth-Anything | SA8295P ADP | SA8295P | QNN | 279.83 ms | 0 - 15 MB | FP16 | NPU | Use Export Script |
|
54 |
+
| Depth-Anything | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 329.299 ms | 1 - 66 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
55 |
+
| Depth-Anything | SA8650 (Proxy) | SA8650P Proxy | QNN | 241.205 ms | 3 - 6 MB | FP16 | NPU | Use Export Script |
|
56 |
+
| Depth-Anything | SA8775P ADP | SA8775P | TFLITE | 368.831 ms | 1 - 270 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
57 |
+
| Depth-Anything | SA8775P ADP | SA8775P | QNN | 263.584 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
|
58 |
+
| Depth-Anything | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 379.814 ms | 1 - 256 MB | FP16 | NPU | [Depth-Anything.tflite](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.tflite) |
|
59 |
+
| Depth-Anything | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 427.625 ms | 3 - 276 MB | FP16 | NPU | Use Export Script |
|
60 |
+
| Depth-Anything | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 207.005 ms | 3 - 3 MB | FP16 | NPU | Use Export Script |
|
61 |
+
| Depth-Anything | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 271.796 ms | 63 - 63 MB | FP16 | NPU | [Depth-Anything.onnx](https://huggingface.co/qualcomm/Depth-Anything/blob/main/Depth-Anything.onnx) |
|
62 |
|
63 |
|
64 |
|
65 |
|
66 |
## Installation
|
67 |
|
|
|
68 |
|
69 |
+
Install the package via pip:
|
70 |
```bash
|
71 |
+
pip install "qai-hub-models[depth-anything]"
|
72 |
```
|
73 |
|
74 |
|
|
|
75 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
76 |
|
77 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
|
|
122 |
Depth-Anything
|
123 |
Device : Samsung Galaxy S23 (13)
|
124 |
Runtime : TFLITE
|
125 |
+
Estimated inference time (ms) : 327.9
|
126 |
+
Estimated peak memory usage (MB): [1, 60]
|
127 |
+
Total # Ops : 634
|
128 |
+
Compute Unit(s) : NPU (634 ops)
|
129 |
```
|
130 |
|
131 |
|
|
|
150 |
torch_model = Model.from_pretrained()
|
151 |
|
152 |
# Device
|
153 |
+
device = hub.Device("Samsung Galaxy S24")
|
154 |
|
155 |
# Trace model
|
156 |
input_shape = torch_model.get_input_spec()
|
|
|
242 |
|
243 |
|
244 |
## License
|
245 |
+
* The license for the original implementation of Depth-Anything can be found
|
246 |
+
[here](https://github.com/huggingface/transformers/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 |
|