User | shimonocyou |
Upload Date | October 20 2024 12:03 PM |
Views | 5 |
AI Information | |
---|---|
Framework | TensorFlow Lite |
Backend | QNN |
Device | Qualcomm Snapdragon 665 |
System Information | |
---|---|
Operating System | Android 12 |
Model | Sony Xperia 10 II |
Model ID | Sony SO-41A |
Motherboard | trinket |
Governor | schedutil |
CPU Information | |
---|---|
Name | Qualcomm Qualcomm |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 81 architecture 8 variant 10 part 2048 revision 2 |
Base Frequency | 1.80 GHz |
Cluster 1 | 4 Cores @ 1.80 GHz |
Cluster 2 | 4 Cores @ 2.02 GHz |
Memory Information | |
---|---|
Size | 3.55 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
41
7.67 IPS |
|
Image Classification (HP)
|
100% |
42
7.79 IPS |
|
Image Classification (Q)
|
99% |
68
12.7 IPS |
|
Image Segmentation (SP)
|
100% |
56
0.90 IPS |
|
Image Segmentation (HP)
|
100% |
56
0.91 IPS |
|
Image Segmentation (Q)
|
98% |
77
1.25 IPS |
|
Pose Estimation (SP)
|
100% |
80
0.09 IPS |
|
Pose Estimation (HP)
|
100% |
80
0.09 IPS |
|
Pose Estimation (Q)
|
98% |
150
0.18 IPS |
|
Object Detection (SP)
|
100% |
41
3.29 IPS |
|
Object Detection (HP)
|
100% |
42
3.34 IPS |
|
Object Detection (Q)
|
87% |
70
5.63 IPS |
|
Face Detection (SP)
|
100% |
110
1.30 IPS |
|
Face Detection (HP)
|
100% |
111
1.32 IPS |
|
Face Detection (Q)
|
97% |
167
1.99 IPS |
|
Depth Estimation (SP)
|
100% |
98
0.75 IPS |
|
Depth Estimation (HP)
|
99% |
98
0.75 IPS |
|
Depth Estimation (Q)
|
64% |
138
1.25 IPS |
|
Style Transfer (SP)
|
100% |
183
0.24 IPS |
|
Style Transfer (HP)
|
100% |
184
0.24 IPS |
|
Style Transfer (Q)
|
98% |
336
0.43 IPS |
|
Image Super-Resolution (SP)
|
100% |
48
1.77 IPS |
|
Image Super-Resolution (HP)
|
100% |
48
1.77 IPS |
|
Image Super-Resolution (Q)
|
97% |
87
3.22 IPS |
|
Text Classification (SP)
|
100% |
61
80.8 IPS |
|
Text Classification (HP)
|
100% |
61
81.0 IPS |
|
Text Classification (Q)
|
91% |
76
101.9 IPS |
|
Machine Translation (SP)
|
100% |
105
1.81 IPS |
|
Machine Translation (HP)
|
100% |
109
1.87 IPS |
|
Machine Translation (Q)
|
57% |
74
1.80 IPS |