User | dudleynyc |
Upload Date | December 01 2024 11:32 PM |
Views | 9 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | ARM |
System Information | |
---|---|
Operating System | iOS 18.1.1 |
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% |
1977
367.6 IPS |
|
Image Classification (HP)
|
100% |
15221
2.83 KIPS |
|
Image Classification (Q)
|
97% |
20886
3.90 KIPS |
|
Image Segmentation (SP)
|
100% |
1914
31.0 IPS |
|
Image Segmentation (HP)
|
100% |
18327
297.1 IPS |
|
Image Segmentation (Q)
|
99% |
24133
391.2 IPS |
|
Pose Estimation (SP)
|
100% |
8417
9.82 IPS |
|
Pose Estimation (HP)
|
100% |
148832
173.7 IPS |
|
Pose Estimation (Q)
|
95% |
287752
337.3 IPS |
|
Object Detection (SP)
|
100% |
2069
164.1 IPS |
|
Object Detection (HP)
|
100% |
16416
1.30 KIPS |
|
Object Detection (Q)
|
91% |
22415
1.79 KIPS |
|
Face Detection (SP)
|
100% |
3463
41.1 IPS |
|
Face Detection (HP)
|
100% |
42109
500.3 IPS |
|
Face Detection (Q)
|
100% |
58478
694.8 IPS |
|
Depth Estimation (SP)
|
100% |
7402
57.0 IPS |
|
Depth Estimation (HP)
|
99% |
95408
735.1 IPS |
|
Depth Estimation (Q)
|
85% |
161509
1.26 KIPS |
|
Style Transfer (SP)
|
100% |
22846
29.4 IPS |
|
Style Transfer (HP)
|
100% |
245177
315.2 IPS |
|
Style Transfer (Q)
|
96% |
306881
396.0 IPS |
|
Image Super-Resolution (SP)
|
100% |
5184
191.4 IPS |
|
Image Super-Resolution (HP)
|
100% |
52556
1.94 KIPS |
|
Image Super-Resolution (Q)
|
99% |
82279
3.04 KIPS |
|
Text Classification (SP)
|
100% |
3413
4.56 KIPS |
|
Text Classification (HP)
|
100% |
4646
6.20 KIPS |
|
Text Classification (Q)
|
93% |
4741
6.36 KIPS |
|
Machine Translation (SP)
|
100% |
3719
64.1 IPS |
|
Machine Translation (HP)
|
100% |
9661
166.4 IPS |
|
Machine Translation (Q)
|
98% |
10807
186.8 IPS |