Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -35,34 +35,33 @@ 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 |
-
| ResNet-Mixed-Convolution | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE |
|
39 |
-
| ResNet-Mixed-Convolution | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 17.
|
40 |
-
| ResNet-Mixed-Convolution | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE |
|
41 |
-
| ResNet-Mixed-Convolution | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 13.
|
42 |
-
| ResNet-Mixed-Convolution | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE |
|
43 |
-
| ResNet-Mixed-Convolution | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 12.
|
44 |
-
| ResNet-Mixed-Convolution | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 166.
|
45 |
-
| ResNet-Mixed-Convolution | SA7255P ADP | SA7255P | TFLITE | 1030.
|
46 |
-
| ResNet-Mixed-Convolution | SA8255 (Proxy) | SA8255P Proxy | TFLITE |
|
47 |
-
| ResNet-Mixed-Convolution | SA8295P ADP | SA8295P | TFLITE |
|
48 |
-
| ResNet-Mixed-Convolution | SA8650 (Proxy) | SA8650P Proxy | TFLITE |
|
49 |
-
| ResNet-Mixed-Convolution | SA8775P ADP | SA8775P | TFLITE | 219.
|
50 |
-
| ResNet-Mixed-Convolution | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE |
|
51 |
-
| ResNet-Mixed-Convolution | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 18.
|
52 |
|
53 |
|
54 |
|
55 |
|
56 |
## Installation
|
57 |
|
58 |
-
This model can be installed as a Python package via pip.
|
59 |
|
|
|
60 |
```bash
|
61 |
-
pip install "qai-hub-models[
|
62 |
```
|
63 |
|
64 |
|
65 |
-
|
66 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
67 |
|
68 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
@@ -113,8 +112,8 @@ Profiling Results
|
|
113 |
ResNet-Mixed-Convolution
|
114 |
Device : Samsung Galaxy S23 (13)
|
115 |
Runtime : TFLITE
|
116 |
-
Estimated inference time (ms) :
|
117 |
-
Estimated peak memory usage (MB): [
|
118 |
Total # Ops : 55
|
119 |
Compute Unit(s) : NPU (50 ops) CPU (5 ops)
|
120 |
```
|
@@ -141,7 +140,7 @@ from qai_hub_models.models.resnet_mixed import Model
|
|
141 |
torch_model = Model.from_pretrained()
|
142 |
|
143 |
# Device
|
144 |
-
device = hub.Device("Samsung Galaxy
|
145 |
|
146 |
# Trace model
|
147 |
input_shape = torch_model.get_input_spec()
|
@@ -219,7 +218,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
|
|
219 |
|
220 |
|
221 |
## License
|
222 |
-
* The license for the original implementation of ResNet-Mixed-Convolution can be found
|
|
|
223 |
* 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)
|
224 |
|
225 |
|
|
|
35 |
|
36 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
37 |
|---|---|---|---|---|---|---|---|---|
|
38 |
+
| ResNet-Mixed-Convolution | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 165.098 ms | 31 - 76 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
39 |
+
| ResNet-Mixed-Convolution | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 17.778 ms | 2 - 73 MB | FP16 | NPU | [ResNet-Mixed-Convolution.onnx](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.onnx) |
|
40 |
+
| ResNet-Mixed-Convolution | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 125.942 ms | 30 - 59 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
41 |
+
| ResNet-Mixed-Convolution | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 13.274 ms | 4 - 68 MB | FP16 | NPU | [ResNet-Mixed-Convolution.onnx](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.onnx) |
|
42 |
+
| ResNet-Mixed-Convolution | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 151.691 ms | 30 - 61 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
43 |
+
| ResNet-Mixed-Convolution | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 12.489 ms | 3 - 67 MB | FP16 | NPU | [ResNet-Mixed-Convolution.onnx](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.onnx) |
|
44 |
+
| ResNet-Mixed-Convolution | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 166.515 ms | 31 - 86 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
45 |
+
| ResNet-Mixed-Convolution | SA7255P ADP | SA7255P | TFLITE | 1030.612 ms | 31 - 58 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
46 |
+
| ResNet-Mixed-Convolution | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 165.831 ms | 31 - 46 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
47 |
+
| ResNet-Mixed-Convolution | SA8295P ADP | SA8295P | TFLITE | 201.377 ms | 31 - 59 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
48 |
+
| ResNet-Mixed-Convolution | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 164.869 ms | 31 - 56 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
49 |
+
| ResNet-Mixed-Convolution | SA8775P ADP | SA8775P | TFLITE | 219.65 ms | 31 - 58 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
50 |
+
| ResNet-Mixed-Convolution | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 193.127 ms | 31 - 60 MB | FP16 | NPU | [ResNet-Mixed-Convolution.tflite](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.tflite) |
|
51 |
+
| ResNet-Mixed-Convolution | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 18.965 ms | 24 - 24 MB | FP16 | NPU | [ResNet-Mixed-Convolution.onnx](https://huggingface.co/qualcomm/ResNet-Mixed-Convolution/blob/main/ResNet-Mixed-Convolution.onnx) |
|
52 |
|
53 |
|
54 |
|
55 |
|
56 |
## Installation
|
57 |
|
|
|
58 |
|
59 |
+
Install the package via pip:
|
60 |
```bash
|
61 |
+
pip install "qai-hub-models[resnet-mixed]"
|
62 |
```
|
63 |
|
64 |
|
|
|
65 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
66 |
|
67 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
|
|
112 |
ResNet-Mixed-Convolution
|
113 |
Device : Samsung Galaxy S23 (13)
|
114 |
Runtime : TFLITE
|
115 |
+
Estimated inference time (ms) : 165.1
|
116 |
+
Estimated peak memory usage (MB): [31, 76]
|
117 |
Total # Ops : 55
|
118 |
Compute Unit(s) : NPU (50 ops) CPU (5 ops)
|
119 |
```
|
|
|
140 |
torch_model = Model.from_pretrained()
|
141 |
|
142 |
# Device
|
143 |
+
device = hub.Device("Samsung Galaxy S24")
|
144 |
|
145 |
# Trace model
|
146 |
input_shape = torch_model.get_input_spec()
|
|
|
218 |
|
219 |
|
220 |
## License
|
221 |
+
* The license for the original implementation of ResNet-Mixed-Convolution can be found
|
222 |
+
[here](https://github.com/pytorch/vision/blob/main/LICENSE).
|
223 |
* 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)
|
224 |
|
225 |
|