Upload Date | July 26 2024 10:02 AM |
Views | 3 |
System Information | |
---|---|
Operating System | Microsoft Windows 11 Pro (64-bit) |
Model | Microsoft Corporation Surface Pro 9 |
Motherboard | Microsoft Corporation Surface Pro 9 |
Power Plan | Balanced |
CPU Information | |
---|---|
Name | Microsoft SQ3 |
Topology | 1 Processor, 8 Cores |
Identifier | ARMv8 (64-bit) Family 8 Model D4B Revision 0 |
Base Frequency | 3.00 GHz |
Cluster 1 | 4 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 8.00 GB |
Inference Information | |
---|---|
Framework | ONNX |
Backend | DirectML |
Device | Qualcomm(R) Adreno(TM) 8cx Gen 3 |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
855
160.1 IPS |
|
Image Classification (F16)
|
100% |
858
160.6 IPS |
|
Image Classification (I8)
|
100% |
544
101.8 IPS |
|
Image Segmentation (F32)
|
100% |
745
12.5 IPS |
|
Image Segmentation (F16)
|
100% |
1059
17.7 IPS |
|
Image Segmentation (I8)
|
98% |
516
8.61 IPS |
|
Pose Estimation (F32)
|
100% |
3631
4.40 IPS |
|
Pose Estimation (F16)
|
100% |
3092
3.74 IPS |
|
Pose Estimation (I8)
|
100% |
3177
3.85 IPS |
|
Object Detection (F32)
|
100% |
618
46.2 IPS |
|
Object Detection (F16)
|
100% |
577
43.1 IPS |
|
Object Detection (I8)
|
71% |
453
33.8 IPS |
|
Face Detection (F32)
|
100% |
1904
22.6 IPS |
|
Face Detection (F16)
|
100% |
2055
24.4 IPS |
|
Face Detection (I8)
|
93% |
1087
12.9 IPS |
|
Depth Estimation (F32)
|
100% |
3979
30.9 IPS |
|
Depth Estimation (F16)
|
100% |
4025
31.2 IPS |
|
Depth Estimation (I8)
|
94% |
2621
20.3 IPS |
|
Style Transfer (F32)
|
100% |
5648
7.43 IPS |
|
Style Transfer (F16)
|
100% |
5709
7.51 IPS |
|
Style Transfer (I8)
|
98% |
4848
6.38 IPS |
|
Image Super-Resolution (F32)
|
100% |
2743
98.0 IPS |
|
Image Super-Resolution (F16)
|
100% |
2647
94.5 IPS |
|
Image Super-Resolution (I8)
|
99% |
1553
55.4 IPS |
|
Text Classification (F32)
|
100% |
189
271.7 IPS |
|
Text Classification (F16)
|
100% |
202
289.7 IPS |
|
Text Classification (I8)
|
98% |
150
216.1 IPS |
|
Machine Translation (F32)
|
100% |
684
12.6 IPS |
|
Machine Translation (F16)
|
100% |
693
12.8 IPS |
|
Machine Translation (I8)
|
65% |
346
6.36 IPS |