Upload Date | November 24 2024 08:28 PM |
Views | 1 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple M1 |
System Information | |
---|---|
Operating System | iOS 18.0 |
Model | iPad Pro (12.9-inch 5th generation) |
Model ID | iPad13,10 |
Motherboard | J523AP |
CPU Information | |
---|---|
Name | Apple M1 |
Topology | 1 Processor, 8 Cores |
Identifier | ARM |
Base Frequency | 3.19 GHz |
Cluster 1 | 4 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 7.35 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1396
259.6 IPS |
|
Image Classification (HP)
|
100% |
6290
1.17 KIPS |
|
Image Classification (Q)
|
96% |
5968
1.11 KIPS |
|
Image Segmentation (SP)
|
100% |
1385
22.5 IPS |
|
Image Segmentation (HP)
|
100% |
6617
107.3 IPS |
|
Image Segmentation (Q)
|
99% |
6755
109.5 IPS |
|
Pose Estimation (SP)
|
100% |
6531
7.62 IPS |
|
Pose Estimation (HP)
|
100% |
68756
80.2 IPS |
|
Pose Estimation (Q)
|
95% |
66096
77.5 IPS |
|
Object Detection (SP)
|
100% |
1511
119.9 IPS |
|
Object Detection (HP)
|
99% |
7242
576.1 IPS |
|
Object Detection (Q)
|
91% |
6572
525.2 IPS |
|
Face Detection (SP)
|
100% |
3040
36.1 IPS |
|
Face Detection (HP)
|
100% |
16184
192.3 IPS |
|
Face Detection (Q)
|
100% |
15147
180.0 IPS |
|
Depth Estimation (SP)
|
100% |
5817
44.8 IPS |
|
Depth Estimation (HP)
|
99% |
47297
364.4 IPS |
|
Depth Estimation (Q)
|
85% |
47442
370.7 IPS |
|
Style Transfer (SP)
|
100% |
17202
22.1 IPS |
|
Style Transfer (HP)
|
100% |
121210
155.8 IPS |
|
Style Transfer (Q)
|
96% |
114992
148.4 IPS |
|
Image Super-Resolution (SP)
|
100% |
3743
138.2 IPS |
|
Image Super-Resolution (HP)
|
100% |
23143
854.6 IPS |
|
Image Super-Resolution (Q)
|
99% |
23846
880.5 IPS |
|
Text Classification (SP)
|
100% |
2288
3.05 KIPS |
|
Text Classification (HP)
|
100% |
3863
5.16 KIPS |
|
Text Classification (Q)
|
93% |
3809
5.11 KIPS |
|
Machine Translation (SP)
|
100% |
2499
43.0 IPS |
|
Machine Translation (HP)
|
100% |
2342
40.3 IPS |
|
Machine Translation (Q)
|
98% |
2338
40.4 IPS |