Upload Date | November 09 2024 08:03 PM |
Views | 2 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple M1 |
System Information | |
---|---|
Operating System | iOS 18.2 |
Model | iPad Pro (12.9-inch 5th generation) |
Model ID | iPad13,8 |
Motherboard | J522AP |
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% |
1389
258.4 IPS |
|
Image Classification (HP)
|
100% |
6207
1.15 KIPS |
|
Image Classification (Q)
|
96% |
5807
1.08 KIPS |
|
Image Segmentation (SP)
|
100% |
1378
22.3 IPS |
|
Image Segmentation (HP)
|
100% |
9777
158.5 IPS |
|
Image Segmentation (Q)
|
99% |
10067
163.2 IPS |
|
Pose Estimation (SP)
|
100% |
6528
7.62 IPS |
|
Pose Estimation (HP)
|
100% |
68604
80.0 IPS |
|
Pose Estimation (Q)
|
95% |
66023
77.4 IPS |
|
Object Detection (SP)
|
100% |
1520
120.5 IPS |
|
Object Detection (HP)
|
99% |
7241
576.0 IPS |
|
Object Detection (Q)
|
91% |
6658
532.0 IPS |
|
Face Detection (SP)
|
100% |
3038
36.1 IPS |
|
Face Detection (HP)
|
100% |
16609
197.3 IPS |
|
Face Detection (Q)
|
100% |
15139
179.9 IPS |
|
Depth Estimation (SP)
|
100% |
5816
44.8 IPS |
|
Depth Estimation (HP)
|
99% |
47029
362.3 IPS |
|
Depth Estimation (Q)
|
85% |
46890
366.4 IPS |
|
Style Transfer (SP)
|
100% |
17151
22.0 IPS |
|
Style Transfer (HP)
|
100% |
120137
154.4 IPS |
|
Style Transfer (Q)
|
96% |
114906
148.3 IPS |
|
Image Super-Resolution (SP)
|
100% |
3786
139.8 IPS |
|
Image Super-Resolution (HP)
|
100% |
22930
846.7 IPS |
|
Image Super-Resolution (Q)
|
99% |
24077
889.1 IPS |
|
Text Classification (SP)
|
100% |
2289
3.06 KIPS |
|
Text Classification (HP)
|
100% |
3966
5.29 KIPS |
|
Text Classification (Q)
|
93% |
3939
5.29 KIPS |
|
Machine Translation (SP)
|
100% |
2485
42.8 IPS |
|
Machine Translation (HP)
|
100% |
3221
55.5 IPS |
|
Machine Translation (Q)
|
98% |
3387
58.5 IPS |