| User | minhle |
| Upload Date | January 28 2025 10:34 AM |
| Views | 7 |
| AI Information | |
|---|---|
| Framework | TensorFlow Lite |
| Backend | CPU |
| Device | Intel Xeon E5-2698 v3 |
| System Information | |
|---|---|
| Operating System | Debian GNU/Linux 12 (bookworm) |
| Model | HUANANZHI |
| Motherboard | HUANANZHI X99-F8D PLUS |
| CPU Information | |
|---|---|
| Name | Intel Xeon E5-2698 v3 |
| Topology | 2 Processors, 32 Cores, 64 Threads |
| Identifier | GenuineIntel Family 6 Model 63 Stepping 2 |
| Base Frequency | 3.60 GHz |
| Cluster 1 | 0 Cores |
| Memory Information | |
|---|---|
| Size | 62.78 GB |
| Workload | Accuracy | Score | |
|---|---|---|---|
|
Image Classification (SP)
|
100% |
234
43.5 IPS |
|
|
Image Classification (HP)
|
100% |
212
39.4 IPS |
|
|
Image Classification (Q)
|
99% |
224
41.8 IPS |
|
|
Image Segmentation (SP)
|
100% |
465
7.54 IPS |
|
|
Image Segmentation (HP)
|
100% |
427
6.92 IPS |
|
|
Image Segmentation (Q)
|
98% |
240
3.90 IPS |
|
|
Pose Estimation (SP)
|
100% |
649
0.76 IPS |
|
|
Pose Estimation (HP)
|
100% |
651
0.76 IPS |
|
|
Pose Estimation (Q)
|
96% |
1118
1.31 IPS |
|
|
Object Detection (SP)
|
100% |
224
17.8 IPS |
|
|
Object Detection (HP)
|
100% |
222
17.6 IPS |
|
|
Object Detection (Q)
|
85% |
252
20.3 IPS |
|
|
Face Detection (SP)
|
100% |
496
5.89 IPS |
|
|
Face Detection (HP)
|
100% |
520
6.18 IPS |
|
|
Face Detection (Q)
|
97% |
507
6.04 IPS |
|
|
Depth Estimation (SP)
|
100% |
454
3.50 IPS |
|
|
Depth Estimation (HP)
|
99% |
446
3.43 IPS |
|
|
Depth Estimation (Q)
|
63% |
432
4.05 IPS |
|
|
Style Transfer (SP)
|
100% |
798
1.03 IPS |
|
|
Style Transfer (HP)
|
100% |
796
1.02 IPS |
|
|
Style Transfer (Q)
|
98% |
1565
2.02 IPS |
|
|
Image Super-Resolution (SP)
|
100% |
282
10.4 IPS |
|
|
Image Super-Resolution (HP)
|
100% |
297
11.0 IPS |
|
|
Image Super-Resolution (Q)
|
97% |
303
11.2 IPS |
|
|
Text Classification (SP)
|
100% |
171
228.8 IPS |
|
|
Text Classification (HP)
|
100% |
118
156.9 IPS |
|
|
Text Classification (Q)
|
92% |
76
102.4 IPS |
|
|
Machine Translation (SP)
|
100% |
458
7.90 IPS |
|
|
Machine Translation (HP)
|
100% |
468
8.05 IPS |
|
|
Machine Translation (Q)
|
58% |
103
2.41 IPS |