Upload Date | May 08 2024 10:06 PM |
Views | 1 |
System Information | |
---|---|
Operating System | Android 14 |
Model | Google Pixel 8a |
Model ID | Google Pixel 8a |
Motherboard | akita |
Governor | sched_pixel |
CPU Information | |
---|---|
Name | ARM ARMv8 |
Topology | 1 Processor, 9 Cores |
Identifier | ARM implementer 65 architecture 8 variant 1 part 3406 revision 0 |
Base Frequency | 1.70 GHz |
Cluster 1 | 4 Cores @ 1.70 GHz |
Cluster 2 | 4 Cores @ 2.37 GHz |
Cluster 3 | 1 Core @ 2.91 GHz |
Memory Information | |
---|---|
Size | 7.40 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | GPU |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
403
75.4 IPS |
|
Image Classification (F16)
|
100% |
458
85.8 IPS |
|
Image Classification (I8)
|
97% |
434
81.2 IPS |
|
Image Segmentation (F32)
|
100% |
579
9.67 IPS |
|
Image Segmentation (F16)
|
100% |
652
10.9 IPS |
|
Image Segmentation (I8)
|
98% |
631
10.5 IPS |
|
Pose Estimation (F32)
|
100% |
455
0.55 IPS |
|
Pose Estimation (F16)
|
100% |
667
0.81 IPS |
|
Pose Estimation (I8)
|
100% |
663
0.80 IPS |
|
Object Detection (F32)
|
100% |
255
19.0 IPS |
|
Object Detection (F16)
|
100% |
279
20.8 IPS |
|
Object Detection (I8)
|
62% |
331
24.7 IPS |
|
Face Detection (F32)
|
100% |
1086
12.9 IPS |
|
Face Detection (F16)
|
98% |
1034
12.3 IPS |
|
Face Detection (I8)
|
86% |
1115
13.3 IPS |
|
Depth Estimation (F32)
|
100% |
859
6.66 IPS |
|
Depth Estimation (F16)
|
100% |
1038
8.05 IPS |
|
Depth Estimation (I8)
|
95% |
1025
7.95 IPS |
|
Style Transfer (F32)
|
100% |
2476
3.26 IPS |
|
Style Transfer (F16)
|
100% |
3062
4.03 IPS |
|
Style Transfer (I8)
|
98% |
3089
4.06 IPS |
|
Image Super-Resolution (F32)
|
100% |
523
18.7 IPS |
|
Image Super-Resolution (F16)
|
100% |
623
22.2 IPS |
|
Image Super-Resolution (I8)
|
98% |
594
21.2 IPS |
|
Text Classification (F32)
|
100% |
393
564.5 IPS |
|
Text Classification (F16)
|
100% |
384
551.5 IPS |
|
Text Classification (I8)
|
92% |
573
823.0 IPS |
|
Machine Translation (F32)
|
100% |
591
10.9 IPS |
|
Machine Translation (F16)
|
100% |
567
10.4 IPS |
|
Machine Translation (I8)
|
64% |
539
9.93 IPS |