Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -39,8 +39,8 @@ More details on model performance across various devices, can be found
|
|
39 |
|
40 |
| Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
41 |
| ---|---|---|---|---|---|---|---|
|
42 |
-
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 1.
|
43 |
-
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 2.
|
44 |
|
45 |
|
46 |
|
@@ -103,8 +103,8 @@ python -m qai_hub_models.models.wideresnet50_quantized.export
|
|
103 |
Profile Job summary of WideResNet50-Quantized
|
104 |
--------------------------------------------------
|
105 |
Device: SA8255 (Proxy) (13)
|
106 |
-
Estimated Inference Time: 2.
|
107 |
-
Estimated Peak Memory Range: 0.01-
|
108 |
Compute Units: NPU (78) | Total (78)
|
109 |
|
110 |
|
|
|
39 |
|
40 |
| Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
41 |
| ---|---|---|---|---|---|---|---|
|
42 |
+
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 1.776 ms | 0 - 2 MB | INT8 | NPU | [WideResNet50-Quantized.tflite](https://huggingface.co/qualcomm/WideResNet50-Quantized/blob/main/WideResNet50-Quantized.tflite)
|
43 |
+
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 2.031 ms | 0 - 361 MB | INT8 | NPU | [WideResNet50-Quantized.so](https://huggingface.co/qualcomm/WideResNet50-Quantized/blob/main/WideResNet50-Quantized.so)
|
44 |
|
45 |
|
46 |
|
|
|
103 |
Profile Job summary of WideResNet50-Quantized
|
104 |
--------------------------------------------------
|
105 |
Device: SA8255 (Proxy) (13)
|
106 |
+
Estimated Inference Time: 2.03 ms
|
107 |
+
Estimated Peak Memory Range: 0.01-413.72 MB
|
108 |
Compute Units: NPU (78) | Total (78)
|
109 |
|
110 |
|