Upload README.md with huggingface_hub

README.md

@@ -18,7 +18,7 @@ tags:
 
 ResNeXt50 is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.
 
-This model is an implementation of ResNeXt50 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py).
+This model is an implementation of ResNeXt50 found [here]({source_repo}).
 This repository provides scripts to run ResNeXt50 on Qualcomm® devices.
 More details on model performance across various devices, can be found
 [here](https://aihub.qualcomm.com/models/resnext50).
@@ -33,15 +33,32 @@ More details on model performance across various devices, can be found
   - Number of parameters: 25.0M
   - Model size: 95.4 MB
 
+| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
+|---|---|---|---|---|---|---|---|---|
+| ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.525 ms | 0 - 3 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.601 ms | 1 - 80 MB | FP16 | NPU | [ResNeXt50.so](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.so) |
+| ResNeXt50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.794 ms | 0 - 2 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
+| ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.967 ms | 0 - 175 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 2.173 ms | 1 - 36 MB | FP16 | NPU | [ResNeXt50.so](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.so) |
+| ResNeXt50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 2.342 ms | 0 - 177 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
+| ResNeXt50 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.503 ms | 0 - 2 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.515 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
+| ResNeXt50 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.48 ms | 0 - 2 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.48 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
+| ResNeXt50 | SA8775 (Proxy) | SA8775P Proxy | TFLITE | 2.501 ms | 0 - 2 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | SA8775 (Proxy) | SA8775P Proxy | QNN | 2.54 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
+| ResNeXt50 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.486 ms | 0 - 2 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.488 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
+| ResNeXt50 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.247 ms | 0 - 113 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 3.368 ms | 1 - 27 MB | FP16 | NPU | Use Export Script |
+| ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.705 ms | 0 - 61 MB | FP16 | NPU | [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) |
+| ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.799 ms | 1 - 39 MB | FP16 | NPU | Use Export Script |
+| ResNeXt50 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.686 ms | 0 - 62 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
+| ResNeXt50 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.669 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
+| ResNeXt50 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.676 ms | 51 - 51 MB | FP16 | NPU | [ResNeXt50.onnx](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.onnx) |
 
 
 
-| Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
-| ---|---|---|---|---|---|---|---|
-| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 2.547 ms | 0 - 2 MB | FP16 | NPU |  [ResNeXt50.tflite](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.tflite) 
-| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 2.603 ms | 1 - 61 MB | FP16 | NPU |  [ResNeXt50.so](https://huggingface.co/qualcomm/ResNeXt50/blob/main/ResNeXt50.so) 
-
-
 
 ## Installation
 
@@ -96,16 +113,16 @@ device. This script does the following:
 ```bash
 python -m qai_hub_models.models.resnext50.export
 ```
-
 ```
-Profile Job summary of ResNeXt50
---------------------------------------------------
-Device: Snapdragon X Elite CRD (11)
-Estimated Inference Time: 2.66 ms
-Estimated Peak Memory Range: 0.57-0.57 MB
-Compute Units: NPU (126) | Total (126)
-
-
+Profiling Results
+------------------------------------------------------------
+ResNeXt50
+Device                          : Samsung Galaxy S23 (13)
+Runtime                         : TFLITE                 
+Estimated inference time (ms)   : 2.5                    
+Estimated peak memory usage (MB): [0, 3]                 
+Total # Ops                     : 79                     
+Compute Unit(s)                 : NPU (79 ops)           
 ```
 
 
@@ -204,15 +221,19 @@ provides instructions on how to use the `.so` shared library  in an Android appl
 Get more details on ResNeXt50's performance across various devices [here](https://aihub.qualcomm.com/models/resnext50).
 Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
 
+
 ## License
-- The license for the original implementation of ResNeXt50 can be found
-  [here](https://github.com/pytorch/vision/blob/main/LICENSE).
-- 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)
+* The license for the original implementation of ResNeXt50 can be found [here](https://github.com/pytorch/vision/blob/main/LICENSE).
+* 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)
+
+
 
 ## References
 * [Aggregated Residual Transformations for Deep Neural Networks](https://arxiv.org/abs/1611.05431)
 * [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py)
 
+
+
 ## Community
 * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
 * For questions or feedback please [reach out to us](mailto:[email protected]).