Upload Date | August 15 2024 01:05 PM |
Views | 2 |
System Information | |
---|---|
Operating System | iOS 17.6.1 |
Model | iPad Pro (12.9-inch 5th generation) |
Model ID | iPad13,8 |
Motherboard | J522AP |
CPU Information | |
---|---|
Name | Apple M1 |
Topology | 1 Processor, 8 Cores |
Identifier | ARM |
Base Frequency | 3.19 GHz |
Cluster 1 | 4 Cores |
Cluster 2 | 4 Cores |
L1 Instruction Cache | 128 KB x 1 |
L1 Data Cache | 64.0 KB x 1 |
L2 Cache | 4.00 MB x 1 |
Memory Information | |
---|---|
Size | 7.31 GB |
Inference Information | |
---|---|
Framework | Core ML |
Backend | GPU |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
1201
224.8 IPS |
|
Image Classification (F16)
|
100% |
1389
260.0 IPS |
|
Image Classification (I8)
|
99% |
1342
251.1 IPS |
|
Image Segmentation (F32)
|
100% |
826
13.8 IPS |
|
Image Segmentation (F16)
|
100% |
850
14.2 IPS |
|
Image Segmentation (I8)
|
100% |
834
13.9 IPS |
|
Pose Estimation (F32)
|
100% |
16462
19.9 IPS |
|
Pose Estimation (F16)
|
100% |
18020
21.8 IPS |
|
Pose Estimation (I8)
|
100% |
17538
21.2 IPS |
|
Object Detection (F32)
|
100% |
1504
112.3 IPS |
|
Object Detection (F16)
|
100% |
1539
114.9 IPS |
|
Object Detection (I8)
|
97% |
1516
113.2 IPS |
|
Face Detection (F32)
|
100% |
5776
68.7 IPS |
|
Face Detection (F16)
|
98% |
7101
84.4 IPS |
|
Face Detection (I8)
|
97% |
7086
84.3 IPS |
|
Depth Estimation (F32)
|
100% |
11984
92.9 IPS |
|
Depth Estimation (F16)
|
100% |
13761
106.7 IPS |
|
Depth Estimation (I8)
|
37% |
11212
86.9 IPS |
|
Style Transfer (F32)
|
100% |
27142
35.7 IPS |
|
Style Transfer (F16)
|
100% |
30835
40.6 IPS |
|
Style Transfer (I8)
|
100% |
34103
44.9 IPS |
|
Image Super-Resolution (F32)
|
100% |
6687
238.8 IPS |
|
Image Super-Resolution (F16)
|
100% |
6588
235.3 IPS |
|
Image Super-Resolution (I8)
|
100% |
6377
227.7 IPS |
|
Text Classification (F32)
|
100% |
1136
1.63 KIPS |
|
Text Classification (F16)
|
100% |
713
1.03 KIPS |
|
Text Classification (I8)
|
96% |
760
1.09 KIPS |
|
Machine Translation (F32)
|
100% |
567
10.4 IPS |
|
Machine Translation (F16)
|
100% |
589
10.8 IPS |
|
Machine Translation (I8)
|
99% |
568
10.5 IPS |