User | luboznaiko |
Upload Date | September 21 2024 06:43 PM |
Views | 10 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | ARM |
System Information | |
---|---|
Operating System | iOS 18.0 |
Model | iPhone17,2 |
Model ID | iPhone17,2 |
Motherboard | D94AP |
CPU Information | |
---|---|
Name | ARM |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 4.03 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 7.47 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1985
369.2 IPS |
|
Image Classification (HP)
|
100% |
15498
2.88 KIPS |
|
Image Classification (Q)
|
97% |
20611
3.85 KIPS |
|
Image Segmentation (SP)
|
100% |
1943
31.5 IPS |
|
Image Segmentation (HP)
|
100% |
14715
238.5 IPS |
|
Image Segmentation (Q)
|
99% |
18279
296.3 IPS |
|
Pose Estimation (SP)
|
100% |
8398
9.80 IPS |
|
Pose Estimation (HP)
|
100% |
153711
179.4 IPS |
|
Pose Estimation (Q)
|
95% |
292235
342.5 IPS |
|
Object Detection (SP)
|
100% |
2094
166.1 IPS |
|
Object Detection (HP)
|
100% |
16230
1.29 KIPS |
|
Object Detection (Q)
|
91% |
22509
1.80 KIPS |
|
Face Detection (SP)
|
100% |
3530
41.9 IPS |
|
Face Detection (HP)
|
100% |
41105
488.4 IPS |
|
Face Detection (Q)
|
100% |
54809
651.3 IPS |
|
Depth Estimation (SP)
|
100% |
7512
57.9 IPS |
|
Depth Estimation (HP)
|
99% |
95471
735.5 IPS |
|
Depth Estimation (Q)
|
85% |
161173
1.26 KIPS |
|
Style Transfer (SP)
|
100% |
23055
29.6 IPS |
|
Style Transfer (HP)
|
100% |
246251
316.6 IPS |
|
Style Transfer (Q)
|
96% |
301686
389.3 IPS |
|
Image Super-Resolution (SP)
|
100% |
5473
202.1 IPS |
|
Image Super-Resolution (HP)
|
100% |
49945
1.84 KIPS |
|
Image Super-Resolution (Q)
|
99% |
84276
3.11 KIPS |
|
Text Classification (SP)
|
100% |
3424
4.57 KIPS |
|
Text Classification (HP)
|
100% |
4368
5.83 KIPS |
|
Text Classification (Q)
|
93% |
4465
5.99 KIPS |
|
Machine Translation (SP)
|
100% |
3793
65.3 IPS |
|
Machine Translation (HP)
|
100% |
9454
162.9 IPS |
|
Machine Translation (Q)
|
98% |
11543
199.5 IPS |