Upload Date | November 15 2024 03:16 PM |
Views | 1 |
System Information | |
---|---|
Operating System | iOS 16.7.10 |
Model | iPad Pro (9.7-inch) |
Model ID | iPad6,4 |
Motherboard | J128AP |
CPU Information | |
---|---|
Name | Apple A9X |
Topology | 1 Processor, 2 Cores |
Identifier | ARM |
Base Frequency | 2.26 GHz |
Cluster 1 | 2 Cores |
L1 Instruction Cache | 64.0 KB x 1 |
L1 Data Cache | 64.0 KB x 1 |
L2 Cache | 3.00 MB x 1 |
Memory Information | |
---|---|
Size | 1.92 GB |
Inference Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
264
49.3 IPS |
|
Image Classification (F16)
|
100% |
195
36.5 IPS |
|
Image Classification (I8)
|
99% |
206
38.5 IPS |
|
Image Segmentation (F32)
|
100% |
166
2.76 IPS |
|
Image Segmentation (F16)
|
100% |
166
2.77 IPS |
|
Image Segmentation (I8)
|
100% |
167
2.78 IPS |
|
Pose Estimation (F32)
|
100% |
1831
2.22 IPS |
|
Pose Estimation (F16)
|
100% |
1779
2.15 IPS |
|
Pose Estimation (I8)
|
100% |
1797
2.18 IPS |
|
Object Detection (F32)
|
100% |
236
17.6 IPS |
|
Object Detection (F16)
|
100% |
220
16.4 IPS |
|
Object Detection (I8)
|
97% |
248
18.5 IPS |
|
Face Detection (F32)
|
100% |
902
10.7 IPS |
|
Face Detection (F16)
|
100% |
910
10.8 IPS |
|
Face Detection (I8)
|
99% |
908
10.8 IPS |
|
Depth Estimation (F32)
|
100% |
1906
14.8 IPS |
|
Depth Estimation (F16)
|
100% |
1949
15.1 IPS |
|
Depth Estimation (I8)
|
99% |
1827
14.2 IPS |
|
Style Transfer (F32)
|
100% |
3671
4.83 IPS |
|
Style Transfer (F16)
|
100% |
3619
4.76 IPS |
|
Style Transfer (I8)
|
100% |
3817
5.02 IPS |
|
Image Super-Resolution (F32)
|
100% |
850
30.4 IPS |
|
Image Super-Resolution (F16)
|
100% |
881
31.5 IPS |
|
Image Super-Resolution (I8)
|
100% |
864
30.9 IPS |
|
Text Classification (F32)
|
100% |
61
88.1 IPS |
|
Text Classification (F16)
|
100% |
94
134.6 IPS |
|
Text Classification (I8)
|
96% |
93
134.3 IPS |
|
Machine Translation (F32)
|
100% |
149
2.74 IPS |
|
Machine Translation (F16)
|
100% |
107
1.96 IPS |
|
Machine Translation (I8)
|
99% |
128
2.35 IPS |