Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -35,48 +35,47 @@ 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 |
-
| SESR-M5-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.
|
39 |
-
| SESR-M5-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 0.
|
40 |
-
| SESR-M5-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 1.
|
41 |
-
| SESR-M5-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.
|
42 |
-
| SESR-M5-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.
|
43 |
-
| SESR-M5-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 0.
|
44 |
-
| SESR-M5-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.
|
45 |
-
| SESR-M5-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.
|
46 |
-
| SESR-M5-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.
|
47 |
-
| SESR-M5-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 3.
|
48 |
-
| SESR-M5-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN |
|
49 |
-
| SESR-M5-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE |
|
50 |
-
| SESR-M5-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.
|
51 |
-
| SESR-M5-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.
|
52 |
-
| SESR-M5-Quantized | SA7255P ADP | SA7255P | TFLITE | 12.
|
53 |
-
| SESR-M5-Quantized | SA7255P ADP | SA7255P | QNN | 10.
|
54 |
-
| SESR-M5-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.
|
55 |
-
| SESR-M5-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.
|
56 |
-
| SESR-M5-Quantized | SA8295P ADP | SA8295P | TFLITE |
|
57 |
-
| SESR-M5-Quantized | SA8295P ADP | SA8295P | QNN | 1.
|
58 |
-
| SESR-M5-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE |
|
59 |
-
| SESR-M5-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.
|
60 |
-
| SESR-M5-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.
|
61 |
-
| SESR-M5-Quantized | SA8775P ADP | SA8775P | QNN | 1.
|
62 |
-
| SESR-M5-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE |
|
63 |
-
| SESR-M5-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 1.
|
64 |
-
| SESR-M5-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.
|
65 |
-
| SESR-M5-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.
|
66 |
|
67 |
|
68 |
|
69 |
|
70 |
## Installation
|
71 |
|
72 |
-
This model can be installed as a Python package via pip.
|
73 |
|
|
|
74 |
```bash
|
75 |
-
pip install "qai-hub-models[
|
76 |
```
|
77 |
|
78 |
|
79 |
-
|
80 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
81 |
|
82 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
@@ -127,8 +126,8 @@ Profiling Results
|
|
127 |
SESR-M5-Quantized
|
128 |
Device : Samsung Galaxy S23 (13)
|
129 |
Runtime : TFLITE
|
130 |
-
Estimated inference time (ms) : 1.
|
131 |
-
Estimated peak memory usage (MB): [
|
132 |
Total # Ops : 27
|
133 |
Compute Unit(s) : NPU (24 ops) CPU (3 ops)
|
134 |
```
|
@@ -155,7 +154,7 @@ from qai_hub_models.models.sesr_m5_quantized import Model
|
|
155 |
torch_model = Model.from_pretrained()
|
156 |
|
157 |
# Device
|
158 |
-
device = hub.Device("Samsung Galaxy
|
159 |
|
160 |
# Trace model
|
161 |
input_shape = torch_model.get_input_spec()
|
@@ -247,7 +246,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
|
|
247 |
|
248 |
|
249 |
## License
|
250 |
-
* The license for the original implementation of SESR-M5-Quantized can be found
|
|
|
251 |
* 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)
|
252 |
|
253 |
|
|
|
35 |
|
36 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
37 |
|---|---|---|---|---|---|---|---|---|
|
38 |
+
| SESR-M5-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.362 ms | 0 - 8 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
39 |
+
| SESR-M5-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 0.973 ms | 0 - 11 MB | INT8 | NPU | [SESR-M5-Quantized.so](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.so) |
|
40 |
+
| SESR-M5-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 1.062 ms | 0 - 9 MB | INT8 | NPU | [SESR-M5-Quantized.onnx](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.onnx) |
|
41 |
+
| SESR-M5-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.11 ms | 2 - 28 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
42 |
+
| SESR-M5-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.706 ms | 0 - 23 MB | INT8 | NPU | [SESR-M5-Quantized.so](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.so) |
|
43 |
+
| SESR-M5-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 0.77 ms | 0 - 29 MB | INT8 | NPU | [SESR-M5-Quantized.onnx](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.onnx) |
|
44 |
+
| SESR-M5-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.664 ms | 0 - 19 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
45 |
+
| SESR-M5-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.728 ms | 0 - 19 MB | INT8 | NPU | Use Export Script |
|
46 |
+
| SESR-M5-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.729 ms | 0 - 21 MB | INT8 | NPU | [SESR-M5-Quantized.onnx](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.onnx) |
|
47 |
+
| SESR-M5-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 3.647 ms | 2 - 18 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
48 |
+
| SESR-M5-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 2.992 ms | 0 - 12 MB | INT8 | NPU | Use Export Script |
|
49 |
+
| SESR-M5-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 22.895 ms | 2 - 4 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
50 |
+
| SESR-M5-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.389 ms | 1 - 8 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
51 |
+
| SESR-M5-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.688 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
|
52 |
+
| SESR-M5-Quantized | SA7255P ADP | SA7255P | TFLITE | 12.772 ms | 2 - 13 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
53 |
+
| SESR-M5-Quantized | SA7255P ADP | SA7255P | QNN | 10.859 ms | 0 - 9 MB | INT8 | NPU | Use Export Script |
|
54 |
+
| SESR-M5-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.59 ms | 0 - 8 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
55 |
+
| SESR-M5-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.695 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
|
56 |
+
| SESR-M5-Quantized | SA8295P ADP | SA8295P | TFLITE | 2.556 ms | 2 - 19 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
57 |
+
| SESR-M5-Quantized | SA8295P ADP | SA8295P | QNN | 1.765 ms | 0 - 14 MB | INT8 | NPU | Use Export Script |
|
58 |
+
| SESR-M5-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.422 ms | 0 - 8 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
59 |
+
| SESR-M5-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.691 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
|
60 |
+
| SESR-M5-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.474 ms | 1 - 12 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
61 |
+
| SESR-M5-Quantized | SA8775P ADP | SA8775P | QNN | 1.309 ms | 0 - 10 MB | INT8 | NPU | Use Export Script |
|
62 |
+
| SESR-M5-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.557 ms | 2 - 22 MB | INT8 | NPU | [SESR-M5-Quantized.tflite](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.tflite) |
|
63 |
+
| SESR-M5-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 1.127 ms | 0 - 17 MB | INT8 | NPU | Use Export Script |
|
64 |
+
| SESR-M5-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.804 ms | 0 - 0 MB | INT8 | NPU | Use Export Script |
|
65 |
+
| SESR-M5-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.222 ms | 3 - 3 MB | INT8 | NPU | [SESR-M5-Quantized.onnx](https://huggingface.co/qualcomm/SESR-M5-Quantized/blob/main/SESR-M5-Quantized.onnx) |
|
66 |
|
67 |
|
68 |
|
69 |
|
70 |
## Installation
|
71 |
|
|
|
72 |
|
73 |
+
Install the package via pip:
|
74 |
```bash
|
75 |
+
pip install "qai-hub-models[sesr-m5-quantized]"
|
76 |
```
|
77 |
|
78 |
|
|
|
79 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
80 |
|
81 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
|
|
126 |
SESR-M5-Quantized
|
127 |
Device : Samsung Galaxy S23 (13)
|
128 |
Runtime : TFLITE
|
129 |
+
Estimated inference time (ms) : 1.4
|
130 |
+
Estimated peak memory usage (MB): [0, 8]
|
131 |
Total # Ops : 27
|
132 |
Compute Unit(s) : NPU (24 ops) CPU (3 ops)
|
133 |
```
|
|
|
154 |
torch_model = Model.from_pretrained()
|
155 |
|
156 |
# Device
|
157 |
+
device = hub.Device("Samsung Galaxy S24")
|
158 |
|
159 |
# Trace model
|
160 |
input_shape = torch_model.get_input_spec()
|
|
|
246 |
|
247 |
|
248 |
## License
|
249 |
+
* The license for the original implementation of SESR-M5-Quantized can be found
|
250 |
+
[here](https://github.com/quic/aimet-model-zoo/blob/develop/LICENSE.pdf).
|
251 |
* 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)
|
252 |
|
253 |
|