Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -37,8 +37,8 @@ More details on model performance across various devices, can be found
|
|
| 37 |
|
| 38 |
| Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
| 39 |
| ---|---|---|---|---|---|---|---|
|
| 40 |
-
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 0.
|
| 41 |
-
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 1.
|
| 42 |
|
| 43 |
|
| 44 |
## Installation
|
|
@@ -46,10 +46,11 @@ More details on model performance across various devices, can be found
|
|
| 46 |
This model can be installed as a Python package via pip.
|
| 47 |
|
| 48 |
```bash
|
| 49 |
-
pip install qai-hub-models
|
| 50 |
```
|
| 51 |
|
| 52 |
|
|
|
|
| 53 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
| 54 |
|
| 55 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
|
@@ -99,8 +100,8 @@ python -m qai_hub_models.models.resnext50_quantized.export
|
|
| 99 |
Profile Job summary of ResNeXt50Quantized
|
| 100 |
--------------------------------------------------
|
| 101 |
Device: Snapdragon X Elite CRD (11)
|
| 102 |
-
Estimated Inference Time: 1.
|
| 103 |
-
Estimated Peak Memory Range: 0.
|
| 104 |
Compute Units: NPU (78) | Total (78)
|
| 105 |
|
| 106 |
|
|
|
|
| 37 |
|
| 38 |
| Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
| 39 |
| ---|---|---|---|---|---|---|---|
|
| 40 |
+
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 0.941 ms | 0 - 2 MB | INT8 | NPU | [ResNeXt50Quantized.tflite](https://huggingface.co/qualcomm/ResNeXt50Quantized/blob/main/ResNeXt50Quantized.tflite)
|
| 41 |
+
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 1.174 ms | 0 - 10 MB | INT8 | NPU | [ResNeXt50Quantized.so](https://huggingface.co/qualcomm/ResNeXt50Quantized/blob/main/ResNeXt50Quantized.so)
|
| 42 |
|
| 43 |
|
| 44 |
## Installation
|
|
|
|
| 46 |
This model can be installed as a Python package via pip.
|
| 47 |
|
| 48 |
```bash
|
| 49 |
+
pip install "qai-hub-models[resnext50_quantized]"
|
| 50 |
```
|
| 51 |
|
| 52 |
|
| 53 |
+
|
| 54 |
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
| 55 |
|
| 56 |
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
|
|
|
|
| 100 |
Profile Job summary of ResNeXt50Quantized
|
| 101 |
--------------------------------------------------
|
| 102 |
Device: Snapdragon X Elite CRD (11)
|
| 103 |
+
Estimated Inference Time: 1.36 ms
|
| 104 |
+
Estimated Peak Memory Range: 0.39-0.39 MB
|
| 105 |
Compute Units: NPU (78) | Total (78)
|
| 106 |
|
| 107 |
|