User | benched |
Upload Date | October 09 2024 01:47 AM |
Views | 16 |
AI Information | |
---|---|
Framework | TensorFlow Lite |
Backend | NNAPI |
Device | ARM Qualcomm |
System Information | |
---|---|
Operating System | Android 13 |
Model | Xiaomi 220333QNY |
Model ID | Xiaomi 220333QNY |
Motherboard | rain |
Governor | schedutil |
CPU Information | |
---|---|
Name | ARM Qualcomm |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 1 part 3337 revision 0 |
Base Frequency | 1.90 GHz |
Cluster 1 | 4 Cores @ 1.90 GHz |
Cluster 2 | 4 Cores @ 2.40 GHz |
Memory Information | |
---|---|
Size | 3.56 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
82
15.3 IPS |
|
Image Classification (HP)
|
100% |
102
18.9 IPS |
|
Image Classification (Q)
|
97% |
451
84.1 IPS |
|
Image Segmentation (SP)
|
100% |
106
1.72 IPS |
|
Image Segmentation (HP)
|
100% |
67
1.09 IPS |
|
Image Segmentation (Q)
|
98% |
546
8.87 IPS |
|
Pose Estimation (SP)
|
100% |
173
0.20 IPS |
|
Pose Estimation (HP)
|
100% |
331
0.39 IPS |
|
Pose Estimation (Q)
|
96% |
3158
3.70 IPS |
|
Object Detection (SP)
|
100% |
53
4.17 IPS |
|
Object Detection (HP)
|
100% |
50
3.95 IPS |
|
Object Detection (Q)
|
85% |
424
34.1 IPS |
|
Face Detection (SP)
|
100% |
208
2.47 IPS |
|
Face Detection (HP)
|
100% |
131
1.56 IPS |
|
Face Detection (Q)
|
97% |
1309
15.6 IPS |
|
Depth Estimation (SP)
|
99% |
143
1.11 IPS |
|
Depth Estimation (HP)
|
99% |
114
0.88 IPS |
|
Depth Estimation (Q)
|
63% |
2057
19.2 IPS |
|
Style Transfer (SP)
|
100% |
276
0.35 IPS |
|
Style Transfer (HP)
|
100% |
211
0.27 IPS |
|
Style Transfer (Q)
|
98% |
1174
1.51 IPS |
|
Image Super-Resolution (SP)
|
100% |
111
4.09 IPS |
|
Image Super-Resolution (HP)
|
100% |
193
7.14 IPS |
|
Image Super-Resolution (Q)
|
97% |
1402
51.9 IPS |
|
Text Classification (SP)
|
71% |
2
2.45 IPS |
|
Text Classification (HP)
|
100% |
72
96.2 IPS |
|
Text Classification (Q)
|
93% |
37
49.4 IPS |
|
Machine Translation (SP)
|
100% |
54
0.93 IPS |
|
Machine Translation (HP)
|
100% |
126
2.18 IPS |
|
Machine Translation (Q)
|
54% |
65
1.84 IPS |