User | thoughtsimple |
Upload Date | November 02 2024 08:33 PM |
Views | 7 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | ARM |
System Information | |
---|---|
Operating System | iOS 18.1 |
Model | iPad16,2 |
Model ID | iPad16,2 |
Motherboard | J411AP |
CPU Information | |
---|---|
Name | ARM |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 3.77 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 7.73 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1737
323.1 IPS |
|
Image Classification (HP)
|
100% |
11559
2.15 KIPS |
|
Image Classification (Q)
|
97% |
15375
2.87 KIPS |
|
Image Segmentation (SP)
|
100% |
1661
26.9 IPS |
|
Image Segmentation (HP)
|
100% |
15905
257.8 IPS |
|
Image Segmentation (Q)
|
99% |
21038
341.0 IPS |
|
Pose Estimation (SP)
|
100% |
7914
9.23 IPS |
|
Pose Estimation (HP)
|
100% |
107694
125.7 IPS |
|
Pose Estimation (Q)
|
95% |
205283
240.6 IPS |
|
Object Detection (SP)
|
100% |
1832
145.3 IPS |
|
Object Detection (HP)
|
100% |
13794
1.09 KIPS |
|
Object Detection (Q)
|
91% |
19302
1.54 KIPS |
|
Face Detection (SP)
|
100% |
3415
40.6 IPS |
|
Face Detection (HP)
|
100% |
35370
420.3 IPS |
|
Face Detection (Q)
|
100% |
51479
611.7 IPS |
|
Depth Estimation (SP)
|
100% |
7094
54.7 IPS |
|
Depth Estimation (HP)
|
99% |
78554
605.2 IPS |
|
Depth Estimation (Q)
|
85% |
127702
998.4 IPS |
|
Style Transfer (SP)
|
100% |
21033
27.0 IPS |
|
Style Transfer (HP)
|
100% |
207782
267.1 IPS |
|
Style Transfer (Q)
|
96% |
252749
326.2 IPS |
|
Image Super-Resolution (SP)
|
100% |
5017
185.2 IPS |
|
Image Super-Resolution (HP)
|
100% |
34423
1.27 KIPS |
|
Image Super-Resolution (Q)
|
99% |
57370
2.12 KIPS |
|
Text Classification (SP)
|
100% |
3206
4.28 KIPS |
|
Text Classification (HP)
|
100% |
3133
4.18 KIPS |
|
Text Classification (Q)
|
93% |
3202
4.30 KIPS |
|
Machine Translation (SP)
|
100% |
3313
57.1 IPS |
|
Machine Translation (HP)
|
100% |
6350
109.4 IPS |
|
Machine Translation (Q)
|
98% |
7871
136.0 IPS |