Upload README.md with huggingface_hub

README.md

@@ -35,19 +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
 |---|---|---|---|---|---|---|---|---|
-| VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 20.106 ms | 0 - 3 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 15.399 ms | 0 - 193 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
-| VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 14.711 ms | 0 - 384 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 11.118 ms | 1 - 150 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
-| VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 13.71 ms | 0 - 208 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 10.634 ms | 1 - 112 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
-| VIT | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 19.818 ms | 0 - 3 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 20.416 ms | 0 - 3 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | SA8775 (Proxy) | SA8775P Proxy | TFLITE | 20.342 ms | 0 - 3 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 19.95 ms | 0 - 3 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | SA8295P ADP | SA8295P | TFLITE | 27.983 ms | 0 - 192 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 25.148 ms | 0 - 365 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
-| VIT | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 21.682 ms | 171 - 171 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
+| VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 17.366 ms | 0 - 41 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 14.796 ms | 0 - 193 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
+| VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 12.003 ms | 101 - 158 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 10.637 ms | 0 - 101 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
+| VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 11.388 ms | 0 - 59 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 10.273 ms | 0 - 73 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
+| VIT | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 16.928 ms | 0 - 47 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | SA7255P ADP | SA7255P | TFLITE | 258.719 ms | 0 - 58 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 17.288 ms | 0 - 41 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | SA8295P ADP | SA8295P | TFLITE | 27.633 ms | 0 - 53 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 17.409 ms | 0 - 44 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | SA8775P ADP | SA8775P | TFLITE | 24.636 ms | 0 - 58 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 24.509 ms | 0 - 54 MB | FP16 | NPU | [VIT.tflite](https://huggingface.co/qualcomm/VIT/blob/main/VIT.tflite) |
+| VIT | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 20.418 ms | 170 - 170 MB | FP16 | NPU | [VIT.onnx](https://huggingface.co/qualcomm/VIT/blob/main/VIT.onnx) |
 
 
 
@@ -111,8 +112,8 @@ Profiling Results
 VIT
 Device                          : Samsung Galaxy S23 (13)
 Runtime                         : TFLITE                 
-Estimated inference time (ms)   : 20.1                   
-Estimated peak memory usage (MB): [0, 3]                 
+Estimated inference time (ms)   : 17.4                   
+Estimated peak memory usage (MB): [0, 41]                
 Total # Ops                     : 1579                   
 Compute Unit(s)                 : NPU (1579 ops)         
 ```
@@ -133,13 +134,29 @@ in memory using the `jit.trace` and then call the `submit_compile_job` API.
 import torch
 
 import qai_hub as hub
-from qai_hub_models.models.vit import 
+from qai_hub_models.models.vit import Model
 
 # Load the model
+torch_model = Model.from_pretrained()
 
 # Device
 device = hub.Device("Samsung Galaxy S23")
 
+# Trace model
+input_shape = torch_model.get_input_spec()
+sample_inputs = torch_model.sample_inputs()
+
+pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
+
+# Compile model on a specific device
+compile_job = hub.submit_compile_job(
+    model=pt_model,
+    device=device,
+    input_specs=torch_model.get_input_spec(),
+)
+
+# Get target model to run on-device
+target_model = compile_job.get_target_model()
 
 ```