Upload README.md with huggingface_hub

README.md

@@ -1,10 +1,10 @@
 ---
 library_name: pytorch
 license: bsd-3-clause
-pipeline_tag: video-classification
 tags:
 - backbone
 - android
+pipeline_tag: video-classification
 
 ---
 
@@ -35,20 +35,20 @@ More details on model performance across various devices, can be found
 
 | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
 |---|---|---|---|---|---|---|---|---|
-| ResNet-3D | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 21.3 ms | 29 - 921 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 16.085 ms | 0 - 218 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
-| ResNet-3D | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 16.105 ms | 29 - 62 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 12.422 ms | 2 - 54 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
-| ResNet-3D | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 14.054 ms | 28 - 62 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 12.295 ms | 2 - 56 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
-| ResNet-3D | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 21.274 ms | 29 - 921 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | SA7255P ADP | SA7255P | TFLITE | 729.974 ms | 29 - 58 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 21.344 ms | 29 - 922 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | SA8295P ADP | SA8295P | TFLITE | 37.968 ms | 29 - 61 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 20.949 ms | 29 - 922 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | SA8775P ADP | SA8775P | TFLITE | 41.233 ms | 29 - 58 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 34.899 ms | 29 - 66 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
-| ResNet-3D | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 18.055 ms | 65 - 65 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
+| ResNet-3D | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 20.711 ms | 31 - 924 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 16.359 ms | 0 - 188 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
+| ResNet-3D | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 15.116 ms | 29 - 64 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 12.545 ms | 2 - 54 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
+| ResNet-3D | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 16.898 ms | 27 - 61 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 10.355 ms | 2 - 56 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
+| ResNet-3D | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 20.465 ms | 29 - 921 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | SA8295P ADP | SA8295P | TFLITE | 36.946 ms | 29 - 56 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 21.114 ms | 29 - 921 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | SA8775P ADP | SA8775P | TFLITE | 41.334 ms | 29 - 56 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 21.329 ms | 29 - 921 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 41.334 ms | 29 - 56 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 35.0 ms | 29 - 62 MB | FP16 | NPU | [ResNet-3D.tflite](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.tflite) |
+| ResNet-3D | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 18.077 ms | 65 - 65 MB | FP16 | NPU | [ResNet-3D.onnx](https://huggingface.co/qualcomm/ResNet-3D/blob/main/ResNet-3D.onnx) |
 
 
 
@@ -112,8 +112,8 @@ Profiling Results
 ResNet-3D
 Device                          : Samsung Galaxy S23 (13) 
 Runtime                         : TFLITE                  
-Estimated inference time (ms)   : 21.3                    
-Estimated peak memory usage (MB): [29, 921]               
+Estimated inference time (ms)   : 20.7                    
+Estimated peak memory usage (MB): [31, 924]               
 Total # Ops                     : 55                      
 Compute Unit(s)                 : NPU (50 ops) CPU (5 ops)
 ```