Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -35,29 +35,30 @@ More details on model performance across various devices, can be found
|
|
35 |
|
36 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
37 |
|---|---|---|---|---|---|---|---|---|
|
38 |
-
| YOLOv11-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 5.
|
39 |
-
| YOLOv11-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 5.
|
40 |
-
| YOLOv11-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX |
|
41 |
-
| YOLOv11-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 3.
|
42 |
-
| YOLOv11-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 3.
|
43 |
-
| YOLOv11-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 4.
|
44 |
| YOLOv11-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 3.175 ms | 0 - 63 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
45 |
-
| YOLOv11-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 3.
|
46 |
-
| YOLOv11-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 3.
|
47 |
-
| YOLOv11-Detection | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 5.
|
48 |
-
| YOLOv11-Detection | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 5.
|
49 |
-
| YOLOv11-Detection |
|
50 |
-
| YOLOv11-Detection |
|
51 |
-
| YOLOv11-Detection |
|
52 |
-
| YOLOv11-Detection |
|
53 |
-
| YOLOv11-Detection | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 5.446 ms | 1 - 3 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
54 |
-
| YOLOv11-Detection | SA8650 (Proxy) | SA8650P Proxy | QNN | 5.423 ms | 5 - 6 MB | FP16 | NPU | Use Export Script |
|
55 |
| YOLOv11-Detection | SA8295P ADP | SA8295P | TFLITE | 10.482 ms | 0 - 59 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
56 |
-
| YOLOv11-Detection | SA8295P ADP | SA8295P | QNN | 9.
|
57 |
-
| YOLOv11-Detection |
|
58 |
-
| YOLOv11-Detection |
|
59 |
-
| YOLOv11-Detection |
|
60 |
-
| YOLOv11-Detection |
|
|
|
|
|
|
|
61 |
|
62 |
|
63 |
|
@@ -122,8 +123,8 @@ Profiling Results
|
|
122 |
YOLOv11-Detection
|
123 |
Device : Samsung Galaxy S23 (13)
|
124 |
Runtime : TFLITE
|
125 |
-
Estimated inference time (ms) : 5.
|
126 |
-
Estimated peak memory usage (MB): [0,
|
127 |
Total # Ops : 382
|
128 |
Compute Unit(s) : NPU (382 ops)
|
129 |
```
|
@@ -144,13 +145,29 @@ in memory using the `jit.trace` and then call the `submit_compile_job` API.
|
|
144 |
import torch
|
145 |
|
146 |
import qai_hub as hub
|
147 |
-
from qai_hub_models.models.yolov11_det import
|
148 |
|
149 |
# Load the model
|
|
|
150 |
|
151 |
# Device
|
152 |
device = hub.Device("Samsung Galaxy S23")
|
153 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
154 |
|
155 |
```
|
156 |
|
|
|
35 |
|
36 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
37 |
|---|---|---|---|---|---|---|---|---|
|
38 |
+
| YOLOv11-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 5.424 ms | 0 - 16 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
39 |
+
| YOLOv11-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 5.569 ms | 5 - 21 MB | FP16 | NPU | [YOLOv11-Detection.so](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.so) |
|
40 |
+
| YOLOv11-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 6.065 ms | 5 - 10 MB | FP16 | NPU | [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) |
|
41 |
+
| YOLOv11-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 3.872 ms | 0 - 40 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
42 |
+
| YOLOv11-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 3.904 ms | 0 - 49 MB | FP16 | NPU | [YOLOv11-Detection.so](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.so) |
|
43 |
+
| YOLOv11-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 4.13 ms | 5 - 123 MB | FP16 | NPU | [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) |
|
44 |
| YOLOv11-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 3.175 ms | 0 - 63 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
45 |
+
| YOLOv11-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 3.854 ms | 5 - 49 MB | FP16 | NPU | Use Export Script |
|
46 |
+
| YOLOv11-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 3.287 ms | 0 - 76 MB | FP16 | NPU | [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) |
|
47 |
+
| YOLOv11-Detection | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 5.435 ms | 0 - 15 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
48 |
+
| YOLOv11-Detection | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 5.389 ms | 5 - 6 MB | FP16 | NPU | Use Export Script |
|
49 |
+
| YOLOv11-Detection | SA7255P ADP | SA7255P | TFLITE | 60.009 ms | 0 - 37 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
50 |
+
| YOLOv11-Detection | SA7255P ADP | SA7255P | QNN | 58.954 ms | 1 - 6 MB | FP16 | NPU | Use Export Script |
|
51 |
+
| YOLOv11-Detection | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 5.436 ms | 0 - 15 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
52 |
+
| YOLOv11-Detection | SA8255 (Proxy) | SA8255P Proxy | QNN | 5.38 ms | 5 - 6 MB | FP16 | NPU | Use Export Script |
|
|
|
|
|
53 |
| YOLOv11-Detection | SA8295P ADP | SA8295P | TFLITE | 10.482 ms | 0 - 59 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
54 |
+
| YOLOv11-Detection | SA8295P ADP | SA8295P | QNN | 9.427 ms | 1 - 6 MB | FP16 | NPU | Use Export Script |
|
55 |
+
| YOLOv11-Detection | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 5.554 ms | 0 - 19 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
56 |
+
| YOLOv11-Detection | SA8650 (Proxy) | SA8650P Proxy | QNN | 5.374 ms | 5 - 6 MB | FP16 | NPU | Use Export Script |
|
57 |
+
| YOLOv11-Detection | SA8775P ADP | SA8775P | TFLITE | 8.213 ms | 0 - 37 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
58 |
+
| YOLOv11-Detection | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 8.825 ms | 0 - 38 MB | FP16 | NPU | [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) |
|
59 |
+
| YOLOv11-Detection | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 8.433 ms | 5 - 43 MB | FP16 | NPU | Use Export Script |
|
60 |
+
| YOLOv11-Detection | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 5.942 ms | 5 - 5 MB | FP16 | NPU | Use Export Script |
|
61 |
+
| YOLOv11-Detection | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.805 ms | 5 - 5 MB | FP16 | NPU | [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) |
|
62 |
|
63 |
|
64 |
|
|
|
123 |
YOLOv11-Detection
|
124 |
Device : Samsung Galaxy S23 (13)
|
125 |
Runtime : TFLITE
|
126 |
+
Estimated inference time (ms) : 5.4
|
127 |
+
Estimated peak memory usage (MB): [0, 16]
|
128 |
Total # Ops : 382
|
129 |
Compute Unit(s) : NPU (382 ops)
|
130 |
```
|
|
|
145 |
import torch
|
146 |
|
147 |
import qai_hub as hub
|
148 |
+
from qai_hub_models.models.yolov11_det import Model
|
149 |
|
150 |
# Load the model
|
151 |
+
torch_model = Model.from_pretrained()
|
152 |
|
153 |
# Device
|
154 |
device = hub.Device("Samsung Galaxy S23")
|
155 |
|
156 |
+
# Trace model
|
157 |
+
input_shape = torch_model.get_input_spec()
|
158 |
+
sample_inputs = torch_model.sample_inputs()
|
159 |
+
|
160 |
+
pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
|
161 |
+
|
162 |
+
# Compile model on a specific device
|
163 |
+
compile_job = hub.submit_compile_job(
|
164 |
+
model=pt_model,
|
165 |
+
device=device,
|
166 |
+
input_specs=torch_model.get_input_spec(),
|
167 |
+
)
|
168 |
+
|
169 |
+
# Get target model to run on-device
|
170 |
+
target_model = compile_job.get_target_model()
|
171 |
|
172 |
```
|
173 |
|