Upload Date | May 09 2024 06:52 AM |
Views | 1 |
System Information | |
---|---|
Operating System | Android 12 |
Model | Lenovo TB328XU |
Model ID | Lenovo TB328XU |
Motherboard | ums512_1h10 |
Governor | schedutil |
CPU Information | |
---|---|
Name | ARM Unisoc |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 3 part 3338 revision 1 |
Base Frequency | 1.82 GHz |
Cluster 1 | 8 Cores @ 1.82 GHz |
Memory Information | |
---|---|
Size | 2.47 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | CPU |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
157
29.4 IPS |
|
Image Classification (F16)
|
100% |
158
29.5 IPS |
|
Image Classification (I8)
|
97% |
262
49.1 IPS |
|
Image Segmentation (F32)
|
100% |
233
3.89 IPS |
|
Image Segmentation (F16)
|
100% |
229
3.82 IPS |
|
Image Segmentation (I8)
|
98% |
285
4.76 IPS |
|
Pose Estimation (F32)
|
100% |
320
0.39 IPS |
|
Pose Estimation (F16)
|
100% |
322
0.39 IPS |
|
Pose Estimation (I8)
|
100% |
1468
1.78 IPS |
|
Object Detection (F32)
|
100% |
170
12.7 IPS |
|
Object Detection (F16)
|
100% |
168
12.6 IPS |
|
Object Detection (I8)
|
61% |
322
24.0 IPS |
|
Face Detection (F32)
|
100% |
367
4.36 IPS |
|
Face Detection (F16)
|
100% |
368
4.37 IPS |
|
Face Detection (I8)
|
86% |
750
8.92 IPS |
|
Depth Estimation (F32)
|
100% |
346
2.69 IPS |
|
Depth Estimation (F16)
|
100% |
344
2.67 IPS |
|
Depth Estimation (I8)
|
95% |
728
5.64 IPS |
|
Style Transfer (F32)
|
100% |
517
0.68 IPS |
|
Style Transfer (F16)
|
100% |
532
0.70 IPS |
|
Style Transfer (I8)
|
98% |
1381
1.82 IPS |
|
Image Super-Resolution (F32)
|
100% |
204
7.27 IPS |
|
Image Super-Resolution (F16)
|
100% |
202
7.20 IPS |
|
Image Super-Resolution (I8)
|
98% |
606
21.6 IPS |
|
Text Classification (F32)
|
100% |
193
277.8 IPS |
|
Text Classification (F16)
|
100% |
190
272.4 IPS |
|
Text Classification (I8)
|
92% |
154
221.0 IPS |
|
Machine Translation (F32)
|
100% |
337
6.21 IPS |
|
Machine Translation (F16)
|
100% |
341
6.28 IPS |
|
Machine Translation (I8)
|
64% |
183
3.36 IPS |