User | Ricochet |
Upload Date | September 08 2024 05:25 PM |
Views | 3 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | ARM |
System Information | |
---|---|
Operating System | iOS 17.6.1 |
Model | iPad16,3 |
Model ID | iPad16,3 |
Motherboard | J717AP |
CPU Information | |
---|---|
Name | ARM |
Topology | 1 Processor, 9 Cores |
Identifier | ARM |
Base Frequency | 4.41 GHz |
Cluster 1 | 3 Cores |
Cluster 2 | 6 Cores |
Memory Information | |
---|---|
Size | 7.28 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1609
299.2 IPS |
|
Image Classification (HP)
|
100% |
9100
1.69 KIPS |
|
Image Classification (Q)
|
97% |
10410
1.94 KIPS |
|
Image Segmentation (SP)
|
100% |
947
15.4 IPS |
|
Image Segmentation (HP)
|
100% |
6843
110.9 IPS |
|
Image Segmentation (Q)
|
99% |
13973
226.5 IPS |
|
Pose Estimation (SP)
|
100% |
9681
11.3 IPS |
|
Pose Estimation (HP)
|
100% |
107026
124.9 IPS |
|
Pose Estimation (Q)
|
95% |
187833
220.2 IPS |
|
Object Detection (SP)
|
100% |
1848
146.6 IPS |
|
Object Detection (HP)
|
100% |
11389
903.3 IPS |
|
Object Detection (Q)
|
93% |
14093
1.12 KIPS |
|
Face Detection (SP)
|
100% |
3079
36.6 IPS |
|
Face Detection (HP)
|
100% |
16089
191.2 IPS |
|
Face Detection (Q)
|
100% |
19503
231.7 IPS |
|
Depth Estimation (SP)
|
100% |
5771
44.5 IPS |
|
Depth Estimation (HP)
|
99% |
62524
481.7 IPS |
|
Depth Estimation (Q)
|
85% |
90401
706.8 IPS |
|
Style Transfer (SP)
|
100% |
15044
19.3 IPS |
|
Style Transfer (HP)
|
100% |
159572
205.1 IPS |
|
Style Transfer (Q)
|
96% |
172888
223.1 IPS |
|
Image Super-Resolution (SP)
|
100% |
4905
181.1 IPS |
|
Image Super-Resolution (HP)
|
100% |
22818
842.6 IPS |
|
Image Super-Resolution (Q)
|
99% |
35026
1.29 KIPS |
|
Text Classification (SP)
|
100% |
1865
2.49 KIPS |
|
Text Classification (HP)
|
71% |
6900
9.92 KIPS |
|
Text Classification (Q)
|
69% |
6783
9.94 KIPS |
|
Machine Translation (SP)
|
100% |
2215
38.1 IPS |
|
Machine Translation (HP)
|
100% |
2989
51.5 IPS |
|
Machine Translation (Q)
|
98% |
3032
52.4 IPS |