v0.30.5

See https://github.com/quic/ai-hub-models/releases/v0.30.5 for changelog.

README.md

@@ -36,13 +36,13 @@ More details on model performance across various devices, can be found
 
 | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
 |---|---|---|---|---|---|---|---|---|
-| NASNet | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 38.061 ms | 0 - 292 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
-| NASNet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 15.255 ms | 0 - 309 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
-| NASNet | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 7.652 ms | 0 - 21 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
-| NASNet | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 10.976 ms | 0 - 291 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
-| NASNet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 7.661 ms | 0 - 20 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
-| NASNet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.585 ms | 0 - 349 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
-| NASNet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 5.335 ms | 0 - 291 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
+| NASNet | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 279.043 ms | 0 - 331 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
+| NASNet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 15.578 ms | 0 - 329 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
+| NASNet | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 7.612 ms | 0 - 29 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
+| NASNet | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 10.727 ms | 0 - 332 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
+| NASNet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 7.623 ms | 0 - 32 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
+| NASNet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.606 ms | 0 - 384 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
+| NASNet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 5.19 ms | 0 - 330 MB | NPU | [NASNet.tflite](https://huggingface.co/qualcomm/NASNet/blob/main/NASNet.tflite) |
 
 
 
@@ -106,8 +106,8 @@ Profiling Results
 NASNet
 Device                          : cs_8275 (ANDROID 14)                 
 Runtime                         : TFLITE                               
-Estimated inference time (ms)   : 38.1                                 
-Estimated peak memory usage (MB): [0, 292]                             
+Estimated inference time (ms)   : 279.0                                
+Estimated peak memory usage (MB): [0, 331]                             
 Total # Ops                     : 776                                  
 Compute Unit(s)                 : npu (776 ops) gpu (0 ops) cpu (0 ops)
 ```
@@ -196,13 +196,13 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
 You can also run the demo on-device.
 
 ```bash
-python -m qai_hub_models.models.nasnet.demo --on-device
+python -m qai_hub_models.models.nasnet.demo --eval-mode on-device
 ```
 
 **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
 environment, please add the following to your cell (instead of the above).
 ```
-%run -m qai_hub_models.models.nasnet.demo -- --on-device
+%run -m qai_hub_models.models.nasnet.demo -- --eval-mode on-device
 ```