Upload README.md with huggingface_hub

README.md

@@ -14,7 +14,7 @@ tags:
 
 DeepLabV3 is designed for semantic segmentation at multiple scales, trained on the COCO dataset. It uses ResNet50 as a backbone.
 
-This model is an implementation of DeepLabV3-ResNet50 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/deeplabv3.py).
+This model is an implementation of DeepLabV3-ResNet50 found [here]({source_repo}).
 This repository provides scripts to run DeepLabV3-ResNet50 on Qualcomm® devices.
 More details on model performance across various devices, can be found
 [here](https://aihub.qualcomm.com/models/deeplabv3_resnet50).
@@ -30,14 +30,18 @@ More details on model performance across various devices, can be found
   - Model size: 151 MB
   - Number of output classes: 21
 
+| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
+|---|---|---|---|---|---|---|---|---|
+| DeepLabV3-ResNet50 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 291.789 ms | 21 - 191 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
+| DeepLabV3-ResNet50 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 225.775 ms | 21 - 43 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
+| DeepLabV3-ResNet50 | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 289.97 ms | 0 - 233 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
+| DeepLabV3-ResNet50 | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 290.802 ms | 0 - 142 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
+| DeepLabV3-ResNet50 | SA8775 (Proxy) | SA8775P Proxy | TFLITE | 289.879 ms | 2 - 139 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
+| DeepLabV3-ResNet50 | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 290.181 ms | 0 - 142 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
+| DeepLabV3-ResNet50 | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 757.728 ms | 21 - 51 MB | FP16 | GPU | [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) |
 
 
 
-| 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 | 291.699 ms | 0 - 142 MB | FP16 | GPU |  [DeepLabV3-ResNet50.tflite](https://huggingface.co/qualcomm/DeepLabV3-ResNet50/blob/main/DeepLabV3-ResNet50.tflite) 
-
-
 
 ## Installation
 
@@ -92,16 +96,16 @@ device. This script does the following:
 ```bash
 python -m qai_hub_models.models.deeplabv3_resnet50.export
 ```
-
 ```
-Profile Job summary of DeepLabV3-ResNet50
---------------------------------------------------
-Device: SA8255 (Proxy) (13)
-Estimated Inference Time: 290.54 ms
-Estimated Peak Memory Range: 0.02-290.81 MB
-Compute Units: GPU (95) | Total (95)
-
-
+Profiling Results
+------------------------------------------------------------
+DeepLabV3-ResNet50
+Device                          : Samsung Galaxy S23 (13)
+Runtime                         : TFLITE                 
+Estimated inference time (ms)   : 291.8                  
+Estimated peak memory usage (MB): [21, 191]              
+Total # Ops                     : 95                     
+Compute Unit(s)                 : GPU (95 ops)           
 ```
 
 
@@ -200,15 +204,19 @@ provides instructions on how to use the `.so` shared library  in an Android appl
 Get more details on DeepLabV3-ResNet50's performance across various devices [here](https://aihub.qualcomm.com/models/deeplabv3_resnet50).
 Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
 
+
 ## License
-- The license for the original implementation of DeepLabV3-ResNet50 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 DeepLabV3-ResNet50 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
 * [Rethinking Atrous Convolution for Semantic Image Segmentation](https://arxiv.org/abs/1706.05587)
 * [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/deeplabv3.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:ai-hub-support@qti.qualcomm.com).