User | ragnarokchan |
Upload Date | August 16 2024 01:18 PM |
Views | 7 |
AI Information | |
---|---|
Framework | Core ML |
Backend | GPU |
Device | Apple M1 |
System Information | |
---|---|
Operating System | iOS 17.6.1 |
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 | 7.31 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1806
335.9 IPS |
|
Image Classification (HP)
|
100% |
1641
305.3 IPS |
|
Image Classification (Q)
|
99% |
1572
293.2 IPS |
|
Image Segmentation (SP)
|
100% |
2632
42.7 IPS |
|
Image Segmentation (HP)
|
100% |
3515
57.0 IPS |
|
Image Segmentation (Q)
|
100% |
3458
56.1 IPS |
|
Pose Estimation (SP)
|
100% |
16370
19.1 IPS |
|
Pose Estimation (HP)
|
100% |
17895
20.9 IPS |
|
Pose Estimation (Q)
|
96% |
15960
18.7 IPS |
|
Object Detection (SP)
|
100% |
2549
202.2 IPS |
|
Object Detection (HP)
|
100% |
2578
204.5 IPS |
|
Object Detection (Q)
|
92% |
2070
165.2 IPS |
|
Face Detection (SP)
|
100% |
6217
73.9 IPS |
|
Face Detection (HP)
|
100% |
7623
90.6 IPS |
|
Face Detection (Q)
|
100% |
5356
63.6 IPS |
|
Depth Estimation (SP)
|
100% |
11924
91.9 IPS |
|
Depth Estimation (HP)
|
99% |
13663
105.3 IPS |
|
Depth Estimation (Q)
|
97% |
13330
103.1 IPS |
|
Style Transfer (SP)
|
100% |
27799
35.7 IPS |
|
Style Transfer (HP)
|
100% |
32903
42.3 IPS |
|
Style Transfer (Q)
|
100% |
33008
42.4 IPS |
|
Image Super-Resolution (SP)
|
100% |
6096
225.1 IPS |
|
Image Super-Resolution (HP)
|
100% |
6948
256.5 IPS |
|
Image Super-Resolution (Q)
|
98% |
5327
197.3 IPS |
|
Text Classification (SP)
|
100% |
1262
1.69 KIPS |
|
Text Classification (HP)
|
100% |
823
1.10 KIPS |
|
Text Classification (Q)
|
93% |
819
1.10 KIPS |
|
Machine Translation (SP)
|
100% |
857
14.8 IPS |
|
Machine Translation (HP)
|
100% |
867
14.9 IPS |
|
Machine Translation (Q)
|
98% |
850
14.7 IPS |