Upload Date | July 26 2024 11:50 PM |
Views | 1 |
System Information | |
---|---|
Operating System | Android 12 |
Model | motorola moto g22 |
Model ID | motorola moto g22 |
Motherboard | p410ae |
CPU Information | |
---|---|
Name | ARM MT6765H |
Topology | 1 Processor, 8 Cores |
Identifier | ARM implementer 65 architecture 8 variant 0 part 3331 revision 4 |
Base Frequency | 2.30 GHz |
Cluster 1 | 4 Cores @ 1.80 GHz |
Cluster 2 | 4 Cores @ 2.30 GHz |
Memory Information | |
---|---|
Size | 3.68 GB |
Inference Information | |
---|---|
Framework | TensorFlow Lite |
Backend | CPU |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
111
20.8 IPS |
|
Image Classification (F16)
|
100% |
114
21.3 IPS |
|
Image Classification (I8)
|
97% |
155
29.0 IPS |
|
Image Segmentation (F32)
|
100% |
158
2.64 IPS |
|
Image Segmentation (F16)
|
100% |
155
2.58 IPS |
|
Image Segmentation (I8)
|
98% |
178
2.97 IPS |
|
Pose Estimation (F32)
|
100% |
201
0.24 IPS |
|
Pose Estimation (F16)
|
100% |
195
0.24 IPS |
|
Pose Estimation (I8)
|
100% |
535
0.65 IPS |
|
Object Detection (F32)
|
100% |
116
8.66 IPS |
|
Object Detection (F16)
|
100% |
116
8.68 IPS |
|
Object Detection (I8)
|
61% |
182
13.6 IPS |
|
Face Detection (F32)
|
100% |
239
2.84 IPS |
|
Face Detection (F16)
|
100% |
238
2.83 IPS |
|
Face Detection (I8)
|
86% |
440
5.23 IPS |
|
Depth Estimation (F32)
|
100% |
232
1.80 IPS |
|
Depth Estimation (F16)
|
100% |
233
1.80 IPS |
|
Depth Estimation (I8)
|
95% |
403
3.12 IPS |
|
Style Transfer (F32)
|
100% |
360
0.47 IPS |
|
Style Transfer (F16)
|
100% |
359
0.47 IPS |
|
Style Transfer (I8)
|
98% |
822
1.08 IPS |
|
Image Super-Resolution (F32)
|
100% |
145
5.19 IPS |
|
Image Super-Resolution (F16)
|
100% |
146
5.22 IPS |
|
Image Super-Resolution (I8)
|
98% |
307
11.0 IPS |
|
Text Classification (F32)
|
100% |
139
199.9 IPS |
|
Text Classification (F16)
|
100% |
142
204.2 IPS |
|
Text Classification (I8)
|
92% |
133
191.6 IPS |
|
Machine Translation (F32)
|
100% |
232
4.28 IPS |
|
Machine Translation (F16)
|
100% |
228
4.19 IPS |
|
Machine Translation (I8)
|
64% |
76
1.40 IPS |