Upload Date | May 15 2022 09:37 PM |
Views | 3 |
System Information | |
---|---|
Operating System | Android 10 |
Model | samsung SM-G965F |
Model ID | samsung SM-G965F |
Motherboard | universal8890 |
Governor | bioshock |
CPU Information | |
---|---|
Name | ARMv8 |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 83 architecture 8 variant 1 part 1 revision 1 |
Base Frequency | 1.59 GHz |
Cluster 1 | 4 Cores @ 1.59 GHz |
Cluster 2 | 4 Cores @ 2.60 GHz |
Memory Information | |
---|---|
Size | 3.70 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | NNAPI |
Device |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
174
18.6 IPS |
|
Image Classification (F16)
|
100% |
165
17.7 IPS |
|
Image Classification (I8)
|
81% |
142
15.2 IPS |
|
Image Segmentation (F32)
|
100% |
229
1.42 IPS |
|
Image Segmentation (F16)
|
100% |
213
1.32 IPS |
|
Image Segmentation (I8)
|
89% |
260
1.61 IPS |
|
Pose Estimation (F32)
|
100% |
89
0.36 IPS |
|
Pose Estimation (F16)
|
100% |
83
0.33 IPS |
|
Pose Estimation (I8)
|
80% |
43
0.17 IPS |
|
Object Detection (F32)
|
100% |
139
4.88 IPS |
|
Object Detection (F16)
|
100% |
117
4.11 IPS |
|
Object Detection (I8)
|
52% |
161
5.65 IPS |
|
Face Detection (F32)
|
100% |
173
1.98 IPS |
|
Face Detection (F16)
|
100% |
173
1.99 IPS |
|
Face Detection (I8)
|
75% |
199
2.28 IPS |
|
Text Classification (F32)
|
100% |
238
94.1 IPS |
|
Text Classification (F16)
|
100% |
238
94.1 IPS |
|
Machine Translation (F32)
|
100% |
97
15.0 IPS |
|
Machine Translation (F16)
|
100% |
97
15.0 IPS |