NVIDIA SHIELD Android TV
Geekbench ML Score
225
TensorFlow Lite GPU Inference Score
Geekbench ML 0.5.0 for Android AArch64
Result Information
| Upload Date | October 23 2021 10:30 PM |
| Views | 3 |
System Information
| System Information | |
|---|---|
| Operating System | Android 9 |
| Model | NVIDIA SHIELD Android TV |
| Model ID | NVIDIA SHIELD Android TV |
| Motherboard | darcy |
| Governor | schedutil |
| CPU Information | |
|---|---|
| Name | ARM ARMv8 |
| Topology | 1 Processor, 4 Cores |
| Identifier | ARM implementer 65 architecture 8 variant 1 part 3335 revision 1 |
| Base Frequency | 2.01 GHz |
| Cluster 1 | 4 Cores @ 2.01 GHz |
| Memory Information | |
|---|---|
| Size | 2.88 GB |
| Inference Information | |
|---|---|
| Framework | TensorFlow Lite |
| Backend | GPU |
| Device | |
Inference Performance
| Workload | Accuracy | Score | |
|---|---|---|---|
|
Image Classification (F32)
|
100% |
389
41.7 IPS |
|
|
Image Classification (F16)
|
100% |
546
58.5 IPS |
|
|
Image Classification (I8)
|
81% |
534
57.2 IPS |
|
|
Image Segmentation (F32)
|
100% |
166
1.03 IPS |
|
|
Image Segmentation (F16)
|
100% |
167
1.03 IPS |
|
|
Image Segmentation (I8)
|
89% |
1008
6.23 IPS |
|
|
Pose Estimation (F32)
|
100% |
102
0.41 IPS |
|
|
Pose Estimation (F16)
|
100% |
166
0.67 IPS |
|
|
Pose Estimation (I8)
|
80% |
165
0.67 IPS |
|
|
Object Detection (F32)
|
100% |
250
8.78 IPS |
|
|
Object Detection (F16)
|
100% |
312
11.0 IPS |
|
|
Object Detection (I8)
|
56% |
308
10.8 IPS |
|
|
Face Detection (F32)
|
100% |
174
1.99 IPS |
|
|
Face Detection (F16)
|
100% |
174
2.00 IPS |
|
|
Face Detection (I8)
|
75% |
777
8.93 IPS |
|
|
Text Classification (F32)
|
100% |
137
54.2 IPS |
|
|
Text Classification (F16)
|
100% |
135
53.2 IPS |
|
|
Machine Translation (F32)
|
100% |
72
11.1 IPS |
|
|
Machine Translation (F16)
|
100% |
75
11.5 IPS |
