Upload Date | September 17 2024 04:37 AM |
Views | 1 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple A15 Bionic |
System Information | |
---|---|
Operating System | iOS 17.6.1 |
Model | iPad mini (6th generation) |
Model ID | iPad14,2 |
Motherboard | J311AP |
CPU Information | |
---|---|
Name | Apple A15 Bionic |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 2.92 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 3.78 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
911
169.4 IPS |
|
Image Classification (HP)
|
100% |
3894
724.2 IPS |
|
Image Classification (Q)
|
96% |
5151
962.0 IPS |
|
Image Segmentation (SP)
|
100% |
602
9.76 IPS |
|
Image Segmentation (HP)
|
100% |
3789
61.4 IPS |
|
Image Segmentation (Q)
|
99% |
5885
95.4 IPS |
|
Pose Estimation (SP)
|
100% |
5279
6.16 IPS |
|
Pose Estimation (HP)
|
100% |
61914
72.2 IPS |
|
Pose Estimation (Q)
|
95% |
67027
78.6 IPS |
|
Object Detection (SP)
|
100% |
954
75.7 IPS |
|
Object Detection (HP)
|
99% |
5334
424.3 IPS |
|
Object Detection (Q)
|
93% |
6070
484.3 IPS |
|
Face Detection (SP)
|
100% |
1620
19.2 IPS |
|
Face Detection (HP)
|
100% |
9666
114.8 IPS |
|
Face Detection (Q)
|
100% |
11322
134.5 IPS |
|
Depth Estimation (SP)
|
100% |
3230
24.9 IPS |
|
Depth Estimation (HP)
|
99% |
34722
267.5 IPS |
|
Depth Estimation (Q)
|
85% |
43381
339.0 IPS |
|
Style Transfer (SP)
|
100% |
9535
12.3 IPS |
|
Style Transfer (HP)
|
100% |
97004
124.7 IPS |
|
Style Transfer (Q)
|
95% |
75771
97.9 IPS |
|
Image Super-Resolution (SP)
|
100% |
2877
106.2 IPS |
|
Image Super-Resolution (HP)
|
100% |
12917
477.0 IPS |
|
Image Super-Resolution (Q)
|
99% |
14878
549.4 IPS |
|
Text Classification (SP)
|
100% |
1005
1.34 KIPS |
|
Text Classification (HP)
|
100% |
2448
3.27 KIPS |
|
Text Classification (Q)
|
93% |
2396
3.22 KIPS |
|
Machine Translation (SP)
|
100% |
1374
23.7 IPS |
|
Machine Translation (HP)
|
100% |
2065
35.6 IPS |
|
Machine Translation (Q)
|
98% |
2051
35.4 IPS |