Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -36,34 +36,38 @@ More details on model performance across various devices, can be found
|
|
36 |
|
37 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
38 |
|---|---|---|---|---|---|---|---|---|
|
39 |
-
| DeepLabV3-Plus-MobileNet
|
40 |
-
| DeepLabV3-Plus-MobileNet
|
41 |
-
| DeepLabV3-Plus-MobileNet
|
42 |
-
| DeepLabV3-Plus-MobileNet
|
43 |
-
| DeepLabV3-Plus-MobileNet
|
44 |
-
| DeepLabV3-Plus-MobileNet
|
45 |
-
| DeepLabV3-Plus-MobileNet
|
46 |
-
| DeepLabV3-Plus-MobileNet
|
47 |
-
| DeepLabV3-Plus-MobileNet
|
48 |
-
| DeepLabV3-Plus-MobileNet
|
49 |
-
| DeepLabV3-Plus-MobileNet
|
50 |
-
| DeepLabV3-Plus-MobileNet
|
51 |
-
| DeepLabV3-Plus-MobileNet
|
52 |
-
| DeepLabV3-Plus-MobileNet
|
53 |
-
| DeepLabV3-Plus-MobileNet
|
54 |
-
| DeepLabV3-Plus-MobileNet
|
55 |
-
| DeepLabV3-Plus-MobileNet
|
56 |
-
| DeepLabV3-Plus-MobileNet
|
57 |
-
| DeepLabV3-Plus-MobileNet
|
58 |
-
| DeepLabV3-Plus-MobileNet
|
59 |
-
| DeepLabV3-Plus-MobileNet
|
60 |
-
| DeepLabV3-Plus-MobileNet
|
61 |
-
| DeepLabV3-Plus-MobileNet
|
62 |
-
| DeepLabV3-Plus-MobileNet
|
63 |
-
| DeepLabV3-Plus-MobileNet
|
64 |
-
| DeepLabV3-Plus-MobileNet
|
65 |
-
| DeepLabV3-Plus-MobileNet
|
66 |
-
| DeepLabV3-Plus-MobileNet
|
|
|
|
|
|
|
|
|
67 |
|
68 |
|
69 |
|
@@ -124,11 +128,11 @@ python -m qai_hub_models.models.deeplabv3_plus_mobilenet_quantized.export
|
|
124 |
```
|
125 |
Profiling Results
|
126 |
------------------------------------------------------------
|
127 |
-
DeepLabV3-Plus-MobileNet
|
128 |
Device : Samsung Galaxy S23 (13)
|
129 |
Runtime : TFLITE
|
130 |
Estimated inference time (ms) : 4.4
|
131 |
-
Estimated peak memory usage (MB): [0,
|
132 |
Total # Ops : 136
|
133 |
Compute Unit(s) : NPU (136 ops)
|
134 |
```
|
@@ -212,20 +216,6 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
|
|
212 |
|
213 |
|
214 |
|
215 |
-
## Run demo on a cloud-hosted device
|
216 |
-
|
217 |
-
You can also run the demo on-device.
|
218 |
-
|
219 |
-
```bash
|
220 |
-
python -m qai_hub_models.models.deeplabv3_plus_mobilenet_quantized.demo --on-device
|
221 |
-
```
|
222 |
-
|
223 |
-
**NOTE**: If you want running in a Jupyter Notebook or Google Colab like
|
224 |
-
environment, please add the following to your cell (instead of the above).
|
225 |
-
```
|
226 |
-
%run -m qai_hub_models.models.deeplabv3_plus_mobilenet_quantized.demo -- --on-device
|
227 |
-
```
|
228 |
-
|
229 |
|
230 |
## Deploying compiled model to Android
|
231 |
|
|
|
36 |
|
37 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
38 |
|---|---|---|---|---|---|---|---|---|
|
39 |
+
| DeepLabV3-Plus-MobileNet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 4.37 ms | 0 - 14 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
40 |
+
| DeepLabV3-Plus-MobileNet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 3.967 ms | 1 - 3 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.so](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.so) |
|
41 |
+
| DeepLabV3-Plus-MobileNet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 69.209 ms | 81 - 150 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.onnx](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.onnx) |
|
42 |
+
| DeepLabV3-Plus-MobileNet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 3.137 ms | 0 - 42 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
43 |
+
| DeepLabV3-Plus-MobileNet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 2.813 ms | 1 - 21 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.so](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.so) |
|
44 |
+
| DeepLabV3-Plus-MobileNet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 53.903 ms | 95 - 249 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.onnx](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.onnx) |
|
45 |
+
| DeepLabV3-Plus-MobileNet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 2.518 ms | 0 - 36 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
46 |
+
| DeepLabV3-Plus-MobileNet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 2.475 ms | 0 - 35 MB | INT8 | NPU | Use Export Script |
|
47 |
+
| DeepLabV3-Plus-MobileNet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 49.262 ms | 81 - 223 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.onnx](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.onnx) |
|
48 |
+
| DeepLabV3-Plus-MobileNet | SA7255P ADP | SA7255P | TFLITE | 55.63 ms | 0 - 30 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
49 |
+
| DeepLabV3-Plus-MobileNet | SA7255P ADP | SA7255P | QNN | 55.401 ms | 1 - 11 MB | INT8 | NPU | Use Export Script |
|
50 |
+
| DeepLabV3-Plus-MobileNet | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 4.352 ms | 0 - 14 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
51 |
+
| DeepLabV3-Plus-MobileNet | SA8255 (Proxy) | SA8255P Proxy | QNN | 3.987 ms | 1 - 4 MB | INT8 | NPU | Use Export Script |
|
52 |
+
| DeepLabV3-Plus-MobileNet | SA8295P ADP | SA8295P | TFLITE | 6.924 ms | 0 - 32 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
53 |
+
| DeepLabV3-Plus-MobileNet | SA8295P ADP | SA8295P | QNN | 6.555 ms | 1 - 19 MB | INT8 | NPU | Use Export Script |
|
54 |
+
| DeepLabV3-Plus-MobileNet | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 4.346 ms | 0 - 13 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
55 |
+
| DeepLabV3-Plus-MobileNet | SA8650 (Proxy) | SA8650P Proxy | QNN | 3.99 ms | 1 - 3 MB | INT8 | NPU | Use Export Script |
|
56 |
+
| DeepLabV3-Plus-MobileNet | SA8775P ADP | SA8775P | TFLITE | 5.951 ms | 0 - 30 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
57 |
+
| DeepLabV3-Plus-MobileNet | SA8775P ADP | SA8775P | QNN | 5.416 ms | 1 - 11 MB | INT8 | NPU | Use Export Script |
|
58 |
+
| DeepLabV3-Plus-MobileNet | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 17.63 ms | 0 - 41 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
59 |
+
| DeepLabV3-Plus-MobileNet | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 17.828 ms | 1 - 15 MB | INT8 | NPU | Use Export Script |
|
60 |
+
| DeepLabV3-Plus-MobileNet | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 169.909 ms | 4 - 6 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
61 |
+
| DeepLabV3-Plus-MobileNet | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 55.63 ms | 0 - 30 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
62 |
+
| DeepLabV3-Plus-MobileNet | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 55.401 ms | 1 - 11 MB | INT8 | NPU | Use Export Script |
|
63 |
+
| DeepLabV3-Plus-MobileNet | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 4.346 ms | 0 - 12 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
64 |
+
| DeepLabV3-Plus-MobileNet | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 3.966 ms | 1 - 4 MB | INT8 | NPU | Use Export Script |
|
65 |
+
| DeepLabV3-Plus-MobileNet | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 5.951 ms | 0 - 30 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
66 |
+
| DeepLabV3-Plus-MobileNet | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 5.416 ms | 1 - 11 MB | INT8 | NPU | Use Export Script |
|
67 |
+
| DeepLabV3-Plus-MobileNet | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 5.344 ms | 0 - 47 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.tflite](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.tflite) |
|
68 |
+
| DeepLabV3-Plus-MobileNet | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 5.439 ms | 1 - 43 MB | INT8 | NPU | Use Export Script |
|
69 |
+
| DeepLabV3-Plus-MobileNet | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 4.39 ms | 1 - 1 MB | INT8 | NPU | Use Export Script |
|
70 |
+
| DeepLabV3-Plus-MobileNet | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 65.849 ms | 133 - 133 MB | INT8 | NPU | [DeepLabV3-Plus-MobileNet-Quantized.onnx](https://huggingface.co/qualcomm/DeepLabV3-Plus-MobileNet-Quantized/blob/main/DeepLabV3-Plus-MobileNet.onnx) |
|
71 |
|
72 |
|
73 |
|
|
|
128 |
```
|
129 |
Profiling Results
|
130 |
------------------------------------------------------------
|
131 |
+
DeepLabV3-Plus-MobileNet
|
132 |
Device : Samsung Galaxy S23 (13)
|
133 |
Runtime : TFLITE
|
134 |
Estimated inference time (ms) : 4.4
|
135 |
+
Estimated peak memory usage (MB): [0, 14]
|
136 |
Total # Ops : 136
|
137 |
Compute Unit(s) : NPU (136 ops)
|
138 |
```
|
|
|
216 |
|
217 |
|
218 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
219 |
|
220 |
## Deploying compiled model to Android
|
221 |
|