Upload Date | October 30 2024 12:42 AM |
Views | 1 |
System Information | |
---|---|
Operating System | Microsoft Windows 10 Pro (64-bit) |
Model | Gigabyte Technology Co., Ltd. X399 AORUS Gaming 7 |
Motherboard | Gigabyte Technology Co., Ltd. X399 AORUS Gaming 7 |
Power Plan | High performance |
CPU Information | |
---|---|
Name | AMD Ryzen Threadripper 1950X |
Topology | 1 Processor, 16 Cores, 32 Threads |
Identifier | AuthenticAMD Family 23 Model 1 Stepping 1 |
Base Frequency | 3.60 GHz |
Cluster 1 | 16 Cores |
L1 Instruction Cache | 64.0 KB x 16 |
L1 Data Cache | 32.0 KB x 16 |
L2 Cache | 512 KB x 16 |
L3 Cache | 8.00 MB x 4 |
Memory Information | |
---|---|
Size | 32.00 GB |
Inference Information | |
---|---|
Framework | ONNX |
Backend | DirectML |
Device | NVIDIA GeForce RTX 3070 |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (F32)
|
100% |
5994
1.12 KIPS |
|
Image Classification (F16)
|
100% |
6018
1.13 KIPS |
|
Image Classification (I8)
|
100% |
4421
827.2 IPS |
|
Image Segmentation (F32)
|
100% |
6050
101.0 IPS |
|
Image Segmentation (F16)
|
100% |
6130
102.4 IPS |
|
Image Segmentation (I8)
|
98% |
5453
91.1 IPS |
|
Pose Estimation (F32)
|
100% |
92597
112.1 IPS |
|
Pose Estimation (F16)
|
100% |
92669
112.2 IPS |
|
Pose Estimation (I8)
|
100% |
74349
90.0 IPS |
|
Object Detection (F32)
|
100% |
2748
205.1 IPS |
|
Object Detection (F16)
|
100% |
2805
209.4 IPS |
|
Object Detection (I8)
|
63% |
2550
190.4 IPS |
|
Face Detection (F32)
|
100% |
14771
175.6 IPS |
|
Face Detection (F16)
|
100% |
14613
173.7 IPS |
|
Face Detection (I8)
|
89% |
11803
140.3 IPS |
|
Depth Estimation (F32)
|
100% |
36827
285.6 IPS |
|
Depth Estimation (F16)
|
100% |
35126
272.4 IPS |
|
Depth Estimation (I8)
|
94% |
26896
208.6 IPS |
|
Style Transfer (F32)
|
100% |
146998
193.4 IPS |
|
Style Transfer (F16)
|
100% |
147103
193.5 IPS |
|
Style Transfer (I8)
|
98% |
116006
152.6 IPS |
|
Image Super-Resolution (F32)
|
100% |
23078
824.2 IPS |
|
Image Super-Resolution (F16)
|
100% |
22865
816.6 IPS |
|
Image Super-Resolution (I8)
|
99% |
14706
525.2 IPS |
|
Text Classification (F32)
|
100% |
1758
2.53 KIPS |
|
Text Classification (F16)
|
100% |
1757
2.53 KIPS |
|
Text Classification (I8)
|
98% |
1026
1.47 KIPS |
|
Machine Translation (F32)
|
100% |
1770
32.6 IPS |
|
Machine Translation (F16)
|
100% |
1772
32.6 IPS |
|
Machine Translation (I8)
|
70% |
1109
20.4 IPS |