Upload Date | April 13 2024 08:37 AM |
Views | 1 |
System Information | |
---|---|
Operating System | Android 13 |
Model | Realme GT Neo |
Model ID | realme RMX3031 |
Motherboard | RM6893 |
CPU Information | |
---|---|
Name | ARM MT6893Z/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 | 3 Cores @ 2.60 GHz |
Cluster 3 | 1 Core @ 3.00 GHz |
Memory Information | |
---|---|
Size | 7.32 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | GPU |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
211
39.4 IPS |
|
Image Classification (F16)
|
100% |
245
45.9 IPS |
|
Image Classification (I8)
|
97% |
255
47.8 IPS |
|
Image Segmentation (F32)
|
100% |
454
7.58 IPS |
|
Image Segmentation (F16)
|
100% |
698
11.7 IPS |
|
Image Segmentation (I8)
|
98% |
645
10.8 IPS |
|
Pose Estimation (F32)
|
100% |
270
0.33 IPS |
|
Pose Estimation (F16)
|
100% |
329
0.40 IPS |
|
Pose Estimation (I8)
|
100% |
328
0.40 IPS |
|
Object Detection (F32)
|
100% |
239
17.8 IPS |
|
Object Detection (F16)
|
100% |
344
25.7 IPS |
|
Object Detection (I8)
|
62% |
348
26.0 IPS |
|
Face Detection (F32)
|
100% |
876
10.4 IPS |
|
Face Detection (F16)
|
98% |
1442
17.1 IPS |
|
Face Detection (I8)
|
86% |
1215
14.4 IPS |
|
Depth Estimation (F32)
|
100% |
518
4.02 IPS |
|
Depth Estimation (F16)
|
100% |
692
5.36 IPS |
|
Depth Estimation (I8)
|
95% |
690
5.35 IPS |
|
Style Transfer (F32)
|
100% |
1146
1.51 IPS |
|
Style Transfer (F16)
|
100% |
1759
2.31 IPS |
|
Style Transfer (I8)
|
98% |
1764
2.32 IPS |
|
Image Super-Resolution (F32)
|
100% |
325
11.6 IPS |
|
Image Super-Resolution (F16)
|
100% |
480
17.2 IPS |
|
Image Super-Resolution (I8)
|
98% |
490
17.5 IPS |
|
Text Classification (F32)
|
100% |
159
228.6 IPS |
|
Text Classification (F16)
|
100% |
142
204.0 IPS |
|
Text Classification (I8)
|
92% |
298
427.7 IPS |
|
Machine Translation (F32)
|
100% |
280
5.16 IPS |
|
Machine Translation (F16)
|
100% |
281
5.17 IPS |
|
Machine Translation (I8)
|
64% |
366
6.73 IPS |