Upload Date | September 17 2024 06:23 PM |
Views | 1 |
AI Information | |
---|---|
Framework | TensorFlow Lite |
Backend | CPU |
Device | MediaTek Dimensity 1100 |
System Information | |
---|---|
Operating System | Android 13 |
Model | Xiaomi Poco X3 GT |
Model ID | Xiaomi 21061110AG |
Motherboard | chopin |
CPU Information | |
---|---|
Name | ARM MT6891Z/CZA |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 1 part 3393 revision 0 |
Base Frequency | 2.00 GHz |
Cluster 1 | 4 Cores @ 2.00 GHz |
Cluster 2 | 4 Cores @ 2.60 GHz |
Memory Information | |
---|---|
Size | 7.30 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
274
50.9 IPS |
|
Image Classification (HP)
|
100% |
295
54.9 IPS |
|
Image Classification (Q)
|
100% |
1028
191.1 IPS |
|
Image Segmentation (SP)
|
100% |
375
6.08 IPS |
|
Image Segmentation (HP)
|
100% |
371
6.01 IPS |
|
Image Segmentation (Q)
|
98% |
1152
18.7 IPS |
|
Pose Estimation (SP)
|
100% |
1470
1.72 IPS |
|
Pose Estimation (HP)
|
100% |
1475
1.72 IPS |
|
Pose Estimation (Q)
|
84% |
2937
3.48 IPS |
|
Object Detection (SP)
|
100% |
290
23.0 IPS |
|
Object Detection (HP)
|
100% |
325
25.7 IPS |
|
Object Detection (Q)
|
83% |
986
79.7 IPS |
|
Face Detection (SP)
|
100% |
876
10.4 IPS |
|
Face Detection (HP)
|
100% |
787
9.35 IPS |
|
Face Detection (Q)
|
95% |
1895
22.6 IPS |
|
Depth Estimation (SP)
|
100% |
981
7.56 IPS |
|
Depth Estimation (HP)
|
99% |
973
7.50 IPS |
|
Depth Estimation (Q)
|
64% |
1893
17.3 IPS |
|
Style Transfer (SP)
|
89% |
2430
3.15 IPS |
|
Style Transfer (HP)
|
89% |
2323
3.02 IPS |
|
Style Transfer (Q)
|
98% |
4950
6.38 IPS |
|
Image Super-Resolution (SP)
|
100% |
522
19.3 IPS |
|
Image Super-Resolution (HP)
|
100% |
491
18.1 IPS |
|
Image Super-Resolution (Q)
|
97% |
1148
42.5 IPS |
|
Text Classification (SP)
|
100% |
255
340.9 IPS |
|
Text Classification (HP)
|
100% |
259
346.0 IPS |
|
Text Classification (Q)
|
88% |
550
741.7 IPS |
|
Machine Translation (SP)
|
100% |
432
7.45 IPS |
|
Machine Translation (HP)
|
100% |
420
7.24 IPS |
|
Machine Translation (Q)
|
50% |
276
9.79 IPS |