User | Stephen_L |
Upload Date | August 26 2024 10:50 AM |
Views | 4 |
System Information | |
---|---|
Operating System | iOS 16.7.8 |
Model | iPad Pro (12.9-inch) |
Model ID | iPad6,7 |
Motherboard | J98aAP |
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 | 3.90 GB |
Inference Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Default |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
318
59.5 IPS |
|
Image Classification (F16)
|
100% |
280
52.5 IPS |
|
Image Classification (I8)
|
99% |
277
51.9 IPS |
|
Image Segmentation (F32)
|
100% |
167
2.78 IPS |
|
Image Segmentation (F16)
|
100% |
166
2.77 IPS |
|
Image Segmentation (I8)
|
100% |
164
2.74 IPS |
|
Pose Estimation (F32)
|
100% |
2010
2.43 IPS |
|
Pose Estimation (F16)
|
100% |
1994
2.42 IPS |
|
Pose Estimation (I8)
|
100% |
2017
2.44 IPS |
|
Object Detection (F32)
|
100% |
371
27.7 IPS |
|
Object Detection (F16)
|
100% |
343
25.6 IPS |
|
Object Detection (I8)
|
97% |
362
27.0 IPS |
|
Face Detection (F32)
|
100% |
1357
16.1 IPS |
|
Face Detection (F16)
|
100% |
1280
15.2 IPS |
|
Face Detection (I8)
|
99% |
1256
14.9 IPS |
|
Depth Estimation (F32)
|
100% |
2166
16.8 IPS |
|
Depth Estimation (F16)
|
100% |
2168
16.8 IPS |
|
Depth Estimation (I8)
|
99% |
2151
16.7 IPS |
|
Style Transfer (F32)
|
100% |
5079
6.68 IPS |
|
Style Transfer (F16)
|
100% |
5031
6.62 IPS |
|
Style Transfer (I8)
|
100% |
5001
6.58 IPS |
|
Image Super-Resolution (F32)
|
100% |
971
34.7 IPS |
|
Image Super-Resolution (F16)
|
100% |
930
33.2 IPS |
|
Image Super-Resolution (I8)
|
100% |
911
32.5 IPS |
|
Text Classification (F32)
|
100% |
55
79.7 IPS |
|
Text Classification (F16)
|
100% |
82
117.1 IPS |
|
Text Classification (I8)
|
96% |
84
121.0 IPS |
|
Machine Translation (F32)
|
100% |
173
3.19 IPS |
|
Machine Translation (F16)
|
100% |
140
2.58 IPS |
|
Machine Translation (I8)
|
99% |
175
3.21 IPS |