User | ju83 |
Upload Date | August 21 2024 09:11 AM |
Views | 4 |
AI Information | |
---|---|
Framework | Core ML |
Backend | GPU |
Device | Apple A12 Bionic |
System Information | |
---|---|
Operating System | iOS 17.6.1 |
Model | iPad Air (3rd generation) |
Model ID | iPad11,3 |
Motherboard | J217AP |
CPU Information | |
---|---|
Name | Apple A12 Bionic |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 2.49 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 2.84 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
213
39.5 IPS |
|
Image Classification (HP)
|
100% |
255
47.5 IPS |
|
Image Classification (Q)
|
99% |
252
47.0 IPS |
|
Image Segmentation (SP)
|
100% |
217
3.51 IPS |
|
Image Segmentation (HP)
|
100% |
248
4.02 IPS |
|
Image Segmentation (Q)
|
100% |
249
4.04 IPS |
|
Pose Estimation (SP)
|
100% |
1639
1.91 IPS |
|
Pose Estimation (HP)
|
100% |
2331
2.72 IPS |
|
Pose Estimation (Q)
|
94% |
653
0.77 IPS |
|
Object Detection (SP)
|
100% |
519
41.2 IPS |
|
Object Detection (HP)
|
100% |
515
40.9 IPS |
|
Object Detection (Q)
|
92% |
159
12.7 IPS |
|
Face Detection (SP)
|
100% |
1890
22.5 IPS |
|
Face Detection (HP)
|
100% |
2232
26.5 IPS |
|
Face Detection (Q)
|
100% |
263
3.12 IPS |
|
Depth Estimation (SP)
|
100% |
1756
13.5 IPS |
|
Depth Estimation (HP)
|
99% |
2516
19.4 IPS |
|
Depth Estimation (Q)
|
97% |
2521
19.5 IPS |
|
Style Transfer (SP)
|
100% |
4512
5.80 IPS |
|
Style Transfer (HP)
|
100% |
5572
7.16 IPS |
|
Style Transfer (Q)
|
100% |
5566
7.15 IPS |
|
Image Super-Resolution (SP)
|
100% |
791
29.2 IPS |
|
Image Super-Resolution (HP)
|
100% |
1055
39.0 IPS |
|
Image Super-Resolution (Q)
|
98% |
290
10.7 IPS |
|
Text Classification (SP)
|
100% |
147
195.9 IPS |
|
Text Classification (HP)
|
100% |
200
266.9 IPS |
|
Text Classification (Q)
|
93% |
196
263.3 IPS |
|
Machine Translation (SP)
|
100% |
189
3.25 IPS |
|
Machine Translation (HP)
|
100% |
155
2.67 IPS |
|
Machine Translation (Q)
|
98% |
155
2.69 IPS |