Upload Date | July 18 2024 09:01 PM |
Views | 1 |
System Information | |
---|---|
Operating System | Android 13 |
Model | Samsung Galaxy S20 |
Model ID | samsung SM-G980F |
Motherboard | exynos990 |
Governor | energy_step |
CPU Information | |
---|---|
Name | ARMv8 |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 83 architecture 8 variant 1 part 4 revision 0 |
Base Frequency | 2.00 GHz |
Cluster 1 | 4 Cores @ 2.00 GHz |
Cluster 2 | 2 Cores @ 2.50 GHz |
Cluster 3 | 2 Cores @ 2.73 GHz |
Memory Information | |
---|---|
Size | 7.27 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | CPU |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
171
31.9 IPS |
|
Image Classification (F16)
|
100% |
170
31.9 IPS |
|
Image Classification (I8)
|
97% |
235
43.9 IPS |
|
Image Segmentation (F32)
|
100% |
202
3.38 IPS |
|
Image Segmentation (F16)
|
100% |
175
2.92 IPS |
|
Image Segmentation (I8)
|
98% |
226
3.78 IPS |
|
Pose Estimation (F32)
|
100% |
489
0.59 IPS |
|
Pose Estimation (F16)
|
100% |
490
0.59 IPS |
|
Pose Estimation (I8)
|
100% |
1896
2.30 IPS |
|
Object Detection (F32)
|
100% |
127
9.50 IPS |
|
Object Detection (F16)
|
100% |
141
10.5 IPS |
|
Object Detection (I8)
|
61% |
148
11.0 IPS |
|
Face Detection (F32)
|
100% |
294
3.49 IPS |
|
Face Detection (F16)
|
100% |
297
3.54 IPS |
|
Face Detection (I8)
|
86% |
335
3.98 IPS |
|
Depth Estimation (F32)
|
100% |
355
2.75 IPS |
|
Depth Estimation (F16)
|
100% |
376
2.92 IPS |
|
Depth Estimation (I8)
|
95% |
434
3.37 IPS |
|
Style Transfer (F32)
|
100% |
915
1.20 IPS |
|
Style Transfer (F16)
|
100% |
920
1.21 IPS |
|
Style Transfer (I8)
|
98% |
1918
2.52 IPS |
|
Image Super-Resolution (F32)
|
100% |
222
7.93 IPS |
|
Image Super-Resolution (F16)
|
100% |
187
6.68 IPS |
|
Image Super-Resolution (I8)
|
98% |
825
29.5 IPS |
|
Text Classification (F32)
|
100% |
253
364.3 IPS |
|
Text Classification (F16)
|
100% |
265
381.3 IPS |
|
Text Classification (I8)
|
92% |
275
394.6 IPS |
|
Machine Translation (F32)
|
100% |
338
6.21 IPS |
|
Machine Translation (F16)
|
100% |
321
5.91 IPS |
|
Machine Translation (I8)
|
64% |
63
1.15 IPS |