Upload Date | August 22 2025 04:22 PM |
Views | 3 |
AI Information | |
---|---|
Framework | TensorFlow Lite |
Backend | NNAPI |
Device | ARM Unisoc |
System Information | |
---|---|
Operating System | Android 13 |
Model | OUKITEL WP23 Pro |
Model ID | OUKITEL WP23 Pro |
Motherboard | ums9230_1h10 |
Governor | schedutil |
CPU Information | |
---|---|
Name | ARM Unisoc |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 3 part 3338 revision 1 |
Base Frequency | 1.61 GHz |
Cluster 1 | 8 Cores @ 1.61 GHz |
Memory Information | |
---|---|
Size | 7.41 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
80
14.8 IPS |
|
Image Classification (HP)
|
100% |
80
14.9 IPS |
|
Image Classification (Q)
|
99% |
104
19.4 IPS |
|
Image Segmentation (SP)
|
100% |
105
1.70 IPS |
|
Image Segmentation (HP)
|
100% |
105
1.70 IPS |
|
Image Segmentation (Q)
|
98% |
146
2.38 IPS |
|
Pose Estimation (SP)
|
100% |
474
0.55 IPS |
|
Pose Estimation (HP)
|
100% |
474
0.55 IPS |
|
Pose Estimation (Q)
|
96% |
433
0.51 IPS |
|
Object Detection (SP)
|
100% |
42
3.30 IPS |
|
Object Detection (HP)
|
100% |
42
3.32 IPS |
|
Object Detection (Q)
|
87% |
78
6.28 IPS |
|
Face Detection (SP)
|
100% |
266
3.17 IPS |
|
Face Detection (HP)
|
100% |
267
3.17 IPS |
|
Face Detection (Q)
|
97% |
206
2.46 IPS |
|
Depth Estimation (SP)
|
100% |
356
2.74 IPS |
|
Depth Estimation (HP)
|
99% |
355
2.73 IPS |
|
Depth Estimation (Q)
|
66% |
332
2.91 IPS |
|
Style Transfer (SP)
|
100% |
360
0.46 IPS |
|
Style Transfer (HP)
|
100% |
188
0.24 IPS |
|
Style Transfer (Q)
|
98% |
559
0.72 IPS |
|
Image Super-Resolution (SP)
|
100% |
209
7.73 IPS |
|
Image Super-Resolution (HP)
|
100% |
210
7.74 IPS |
|
Image Super-Resolution (Q)
|
97% |
213
7.90 IPS |
|
Text Classification (SP)
|
100% |
25
33.5 IPS |
|
Text Classification (HP)
|
100% |
59
78.8 IPS |
|
Text Classification (Q)
|
93% |
41
55.0 IPS |
|
Machine Translation (SP)
|
100% |
71
1.23 IPS |
|
Machine Translation (HP)
|
100% |
104
1.80 IPS |
|
Machine Translation (Q)
|
55% |
47
1.24 IPS |