Upload Date | January 19 2022 07:04 AM |
Views | 2 |
System Information | |
---|---|
Operating System | Android 12 |
Model | samsung SM-G9900 |
Model ID | samsung SM-G9900 |
Motherboard | lahaina |
Governor | schedutil |
CPU Information | |
---|---|
Name | ARM SM8350 |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 1 part 3396 revision 0 |
Base Frequency | 1.80 GHz |
Cluster 1 | 4 Cores @ 1.80 GHz |
Cluster 2 | 3 Cores @ 2.42 GHz |
Cluster 3 | 1 Core @ 2.84 GHz |
Memory Information | |
---|---|
Size | 7.19 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | GPU |
Device |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
1321
141.6 IPS |
|
Image Classification (F16)
|
100% |
2041
218.7 IPS |
|
Image Classification (I8)
|
80% |
1852
198.4 IPS |
|
Image Segmentation (F32)
|
100% |
3264
20.2 IPS |
|
Image Segmentation (F16)
|
100% |
5350
33.0 IPS |
|
Image Segmentation (I8)
|
89% |
4877
30.1 IPS |
|
Pose Estimation (F32)
|
100% |
1574
6.37 IPS |
|
Pose Estimation (F16)
|
100% |
2929
11.8 IPS |
|
Pose Estimation (I8)
|
80% |
2904
11.8 IPS |
|
Object Detection (F32)
|
100% |
661
23.2 IPS |
|
Object Detection (F16)
|
99% |
763
26.8 IPS |
|
Object Detection (I8)
|
53% |
784
27.5 IPS |
|
Face Detection (F32)
|
100% |
2491
28.6 IPS |
|
Face Detection (F16)
|
97% |
4170
47.9 IPS |
|
Face Detection (I8)
|
75% |
3643
41.9 IPS |
|
Text Classification (F32)
|
100% |
753
297.7 IPS |
|
Text Classification (F16)
|
98% |
957
378.4 IPS |
|
Machine Translation (F32)
|
100% |
627
96.7 IPS |
|
Machine Translation (F16)
|
100% |
627
96.6 IPS |