Upload Date | May 26 2024 05:44 PM |
Views | 1 |
System Information | |
---|---|
Operating System | Android 14 |
Model | Samsung Galaxy S21 5G |
Model ID | samsung SM-G991W |
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 | NNAPI |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
571
106.8 IPS |
|
Image Classification (F16)
|
100% |
1129
211.2 IPS |
|
Image Classification (I8)
|
94% |
18
3.31 IPS |
|
Image Segmentation (F32)
|
100% |
979
16.3 IPS |
|
Image Segmentation (F16)
|
100% |
1548
25.8 IPS |
|
Image Segmentation (I8)
|
98% |
505
8.44 IPS |
|
Pose Estimation (F32)
|
100% |
1932
2.34 IPS |
|
Pose Estimation (F16)
|
100% |
6889
8.34 IPS |
|
Pose Estimation (I8)
|
100% |
1194
1.45 IPS |
|
Object Detection (F32)
|
100% |
572
42.7 IPS |
|
Object Detection (F16)
|
100% |
958
71.5 IPS |
|
Object Detection (I8)
|
0% |
2805
209.4 IPS |
|
Face Detection (F32)
|
100% |
1635
19.4 IPS |
|
Face Detection (F16)
|
97% |
2669
31.7 IPS |
|
Face Detection (I8)
|
86% |
1174
14.0 IPS |
|
Depth Estimation (F32)
|
100% |
690
5.35 IPS |
|
Depth Estimation (F16)
|
100% |
836
6.48 IPS |
|
Depth Estimation (I8)
|
74% |
7966
61.8 IPS |
|
Style Transfer (F32)
|
100% |
2098
2.76 IPS |
|
Style Transfer (F16)
|
100% |
2743
3.61 IPS |
|
Style Transfer (I8)
|
19% |
578
0.76 IPS |
|
Image Super-Resolution (F32)
|
100% |
1054
37.6 IPS |
|
Image Super-Resolution (F16)
|
100% |
3481
124.3 IPS |
|
Image Super-Resolution (I8)
|
98% |
269
9.62 IPS |
|
Text Classification (F32)
|
100% |
4
6.41 IPS |
|
Text Classification (F16)
|
100% |
60
86.1 IPS |
|
Text Classification (I8)
|
94% |
31
43.9 IPS |
|
Machine Translation (F32)
|
100% |
266
4.89 IPS |
|
Machine Translation (F16)
|
100% |
329
6.05 IPS |
|
Machine Translation (I8)
|
64% |
351
6.46 IPS |