Upload Date | November 16 2024 08:49 PM |
Views | 12 |
AI Information | |
---|---|
Framework | TensorFlow Lite |
Backend | CPU |
Device | ARM ARMv8 |
System Information | |
---|---|
Operating System | Android 13 |
Model | Google kukui |
Model ID | Google kukui |
Motherboard | krane |
Governor | schedutil |
CPU Information | |
---|---|
Name | ARM ARMv8 |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 0 part 3337 revision 2 |
Base Frequency | 1.99 GHz |
Cluster 1 | 8 Cores @ 1.99 GHz |
Memory Information | |
---|---|
Size | 3.13 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
139
25.8 IPS |
|
Image Classification (HP)
|
100% |
148
27.6 IPS |
|
Image Classification (Q)
|
100% |
230
42.8 IPS |
|
Image Segmentation (SP)
|
100% |
193
3.13 IPS |
|
Image Segmentation (HP)
|
100% |
193
3.13 IPS |
|
Image Segmentation (Q)
|
98% |
320
5.20 IPS |
|
Pose Estimation (SP)
|
100% |
702
0.82 IPS |
|
Pose Estimation (HP)
|
100% |
708
0.83 IPS |
|
Pose Estimation (Q)
|
84% |
534
0.63 IPS |
|
Object Detection (SP)
|
100% |
150
11.9 IPS |
|
Object Detection (HP)
|
100% |
148
11.8 IPS |
|
Object Detection (Q)
|
83% |
231
18.6 IPS |
|
Face Detection (SP)
|
100% |
358
4.25 IPS |
|
Face Detection (HP)
|
100% |
379
4.50 IPS |
|
Face Detection (Q)
|
95% |
581
6.94 IPS |
|
Depth Estimation (SP)
|
100% |
499
3.84 IPS |
|
Depth Estimation (HP)
|
99% |
478
3.68 IPS |
|
Depth Estimation (Q)
|
64% |
532
4.88 IPS |
|
Style Transfer (SP)
|
89% |
1260
1.64 IPS |
|
Style Transfer (HP)
|
89% |
1325
1.72 IPS |
|
Style Transfer (Q)
|
98% |
1514
1.95 IPS |
|
Image Super-Resolution (SP)
|
100% |
270
9.98 IPS |
|
Image Super-Resolution (HP)
|
100% |
266
9.83 IPS |
|
Image Super-Resolution (Q)
|
97% |
328
12.1 IPS |
|
Text Classification (SP)
|
100% |
146
195.3 IPS |
|
Text Classification (HP)
|
100% |
149
198.7 IPS |
|
Text Classification (Q)
|
88% |
170
229.6 IPS |
|
Machine Translation (SP)
|
100% |
240
4.13 IPS |
|
Machine Translation (HP)
|
100% |
231
3.98 IPS |
|
Machine Translation (Q)
|
50% |
96
3.42 IPS |