User | danbedford |
Upload Date | November 05 2024 12:49 PM |
Views | 8 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | ARM |
System Information | |
---|---|
Operating System | iOS 18.1 |
Model | iPhone17,2 |
Model ID | iPhone17,2 |
Motherboard | D94AP |
CPU Information | |
---|---|
Name | ARM |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 4.04 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 7.47 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1956
363.8 IPS |
|
Image Classification (HP)
|
100% |
15477
2.88 KIPS |
|
Image Classification (Q)
|
97% |
21042
3.93 KIPS |
|
Image Segmentation (SP)
|
100% |
1850
30.0 IPS |
|
Image Segmentation (HP)
|
100% |
18136
294.0 IPS |
|
Image Segmentation (Q)
|
99% |
23871
387.0 IPS |
|
Pose Estimation (SP)
|
100% |
8285
9.67 IPS |
|
Pose Estimation (HP)
|
100% |
150444
175.5 IPS |
|
Pose Estimation (Q)
|
95% |
293316
343.8 IPS |
|
Object Detection (SP)
|
100% |
2046
162.3 IPS |
|
Object Detection (HP)
|
100% |
16353
1.30 KIPS |
|
Object Detection (Q)
|
91% |
22689
1.81 KIPS |
|
Face Detection (SP)
|
100% |
3356
39.9 IPS |
|
Face Detection (HP)
|
100% |
42169
501.1 IPS |
|
Face Detection (Q)
|
100% |
58620
696.5 IPS |
|
Depth Estimation (SP)
|
100% |
7498
57.8 IPS |
|
Depth Estimation (HP)
|
99% |
95292
734.2 IPS |
|
Depth Estimation (Q)
|
85% |
162622
1.27 KIPS |
|
Style Transfer (SP)
|
100% |
22430
28.8 IPS |
|
Style Transfer (HP)
|
100% |
244495
314.3 IPS |
|
Style Transfer (Q)
|
96% |
303142
391.2 IPS |
|
Image Super-Resolution (SP)
|
100% |
5351
197.6 IPS |
|
Image Super-Resolution (HP)
|
100% |
50210
1.85 KIPS |
|
Image Super-Resolution (Q)
|
99% |
84815
3.13 KIPS |
|
Text Classification (SP)
|
100% |
3452
4.61 KIPS |
|
Text Classification (HP)
|
100% |
4920
6.57 KIPS |
|
Text Classification (Q)
|
93% |
4854
6.51 KIPS |
|
Machine Translation (SP)
|
100% |
3678
63.4 IPS |
|
Machine Translation (HP)
|
100% |
10012
172.5 IPS |
|
Machine Translation (Q)
|
98% |
11416
197.3 IPS |