Upload Date | July 26 2024 02:41 PM |
Views | 1 |
System Information | |
---|---|
Operating System | Android 14 |
Model | Xiaomi 12T Pro |
Model ID | Xiaomi 22081212UG |
Motherboard | diting |
Governor | walt |
CPU Information | |
---|---|
Name | ARM ARMv8 |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 2 part 3400 revision 0 |
Base Frequency | 2.02 GHz |
Cluster 1 | 4 Cores @ 2.02 GHz |
Cluster 2 | 3 Cores @ 2.75 GHz |
Cluster 3 | 1 Core @ 3.19 GHz |
Memory Information | |
---|---|
Size | 10.93 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | CPU |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
339
63.5 IPS |
|
Image Classification (F16)
|
100% |
345
64.5 IPS |
|
Image Classification (I8)
|
97% |
543
101.7 IPS |
|
Image Segmentation (F32)
|
100% |
451
7.53 IPS |
|
Image Segmentation (F16)
|
100% |
451
7.53 IPS |
|
Image Segmentation (I8)
|
98% |
516
8.62 IPS |
|
Pose Estimation (F32)
|
100% |
856
1.04 IPS |
|
Pose Estimation (F16)
|
100% |
868
1.05 IPS |
|
Pose Estimation (I8)
|
100% |
3311
4.01 IPS |
|
Object Detection (F32)
|
100% |
351
26.2 IPS |
|
Object Detection (F16)
|
100% |
342
25.5 IPS |
|
Object Detection (I8)
|
61% |
684
51.0 IPS |
|
Face Detection (F32)
|
100% |
788
9.37 IPS |
|
Face Detection (F16)
|
100% |
781
9.29 IPS |
|
Face Detection (I8)
|
86% |
1345
16.0 IPS |
|
Depth Estimation (F32)
|
100% |
807
6.25 IPS |
|
Depth Estimation (F16)
|
100% |
802
6.22 IPS |
|
Depth Estimation (I8)
|
95% |
1487
11.5 IPS |
|
Style Transfer (F32)
|
100% |
1530
2.01 IPS |
|
Style Transfer (F16)
|
100% |
1558
2.05 IPS |
|
Style Transfer (I8)
|
98% |
2966
3.90 IPS |
|
Image Super-Resolution (F32)
|
100% |
489
17.5 IPS |
|
Image Super-Resolution (F16)
|
100% |
495
17.7 IPS |
|
Image Super-Resolution (I8)
|
98% |
1285
45.9 IPS |
|
Text Classification (F32)
|
100% |
368
528.8 IPS |
|
Text Classification (F16)
|
100% |
386
555.2 IPS |
|
Text Classification (I8)
|
92% |
250
359.6 IPS |
|
Machine Translation (F32)
|
100% |
739
13.6 IPS |
|
Machine Translation (F16)
|
100% |
741
13.6 IPS |
|
Machine Translation (I8)
|
64% |
293
5.40 IPS |