RTX 4090 FE Results

[ConfusedWizard]

* TF Version: 2.10.0
* Platform: Windows-10-10.0.19045-SP0
* CPU: N/A
* CPU RAM: 32 GB
* GPU/0: NVIDIA GeForce RTX 4090
* GPU RAM: 20.8 GB
* CUDA Version: 11.7
* CUDA Build: V11.7.64

Summary:
Device Inference Score: 28423
Device Training Score: 35100
Device AI Score: 63523

Detail:

1/19. MobileNet-V2

1.1 - inference | batch=50, size=224x224: 26.4 ± 3.0 ms
1.2 - training | batch=50, size=224x224: 60.5 ± 12.5 ms

2/19. Inception-V3

2.1 - inference | batch=20, size=346x346: 31.6 ± 3.1 ms
2.2 - training | batch=20, size=346x346: 57.2 ± 2.3 ms

3/19. Inception-V4

3.1 - inference | batch=10, size=346x346: 24.9 ± 0.8 ms
3.2 - training | batch=10, size=346x346: 62.6 ± 4.5 ms

4/19. Inception-ResNet-V2

4.1 - inference | batch=10, size=346x346: 33.9 ± 2.8 ms
4.2 - training | batch=8, size=346x346: 66.9 ± 3.8 ms

5/19. ResNet-V2-50

5.1 - inference | batch=10, size=346x346: 14.9 ± 3.0 ms
5.2 - training | batch=10, size=346x346: 31.8 ± 0.7 ms

6/19. ResNet-V2-152

6.1 - inference | batch=10, size=256x256: 29.2 ± 4.2 ms
6.2 - training | batch=10, size=256x256: 49.0 ± 2.6 ms

7/19. VGG-16

7.1 - inference | batch=20, size=224x224: 24.4 ± 0.9 ms
7.2 - training | batch=2, size=224x224: 18.9 ± 0.6 ms

8/19. SRCNN 9-5-5

8.1 - inference | batch=10, size=512x512: 33.2 ± 8.7 ms
8.2 - inference | batch=1, size=1536x1536: 23.4 ± 2.4 ms
8.3 - training | batch=10, size=512x512: 91.1 ± 13.8 ms

9/19. VGG-19 Super-Res

9.1 - inference | batch=10, size=256x256: 37.2 ± 7.1 ms
9.2 - inference | batch=1, size=1024x1024: 52.1 ± 0.5 ms
9.3 - training | batch=10, size=224x224: 87.9 ± 2.6 ms

10/19. ResNet-SRGAN

10.1 - inference | batch=10, size=512x512: 51.8 ± 9.6 ms
10.2 - inference | batch=1, size=1536x1536: 35.5 ± 0.9 ms
10.3 - training | batch=5, size=512x512: 76.4 ± 16.4 ms

11/19. ResNet-DPED

11.1 - inference | batch=10, size=256x256: 49.5 ± 1.2 ms
11.2 - inference | batch=1, size=1024x1024: 76.9 ± 0.6 ms
11.3 - training | batch=15, size=128x128: 78.8 ± 20.7 ms

12/19. U-Net

12.1 - inference | batch=4, size=512x512: 74.8 ± 0.4 ms
12.2 - inference | batch=1, size=1024x1024: 74.1 ± 1.1 ms
12.3 - training | batch=4, size=256x256: 64.5 ± 0.6 ms

13/19. Nvidia-SPADE

13.1 - inference | batch=5, size=128x128: 26.4 ± 0.6 ms
13.2 - training | batch=1, size=128x128: 44.0 ± 0.4 ms

14/19. ICNet

14.1 - inference | batch=5, size=1024x1536: 56.8 ± 12.0 ms
14.2 - training | batch=10, size=1024x1536: 176 ± 37 ms

15/19. PSPNet

15.1 - inference | batch=5, size=720x720: 102.0 ± 0.4 ms
15.2 - training | batch=1, size=512x512: 37.9 ± 0.5 ms

16/19. DeepLab

16.1 - inference | batch=2, size=512x512: 27.0 ± 0.4 ms
16.2 - training | batch=1, size=384x384: 36.9 ± 0.7 ms

17/19. Pixel-RNN

17.1 - inference | batch=50, size=64x64: 569 ± 36 ms
17.2 - training | batch=10, size=64x64: 2796 ± 10 ms

18/19. LSTM-Sentiment

18.1 - inference | batch=100, size=1024x300: 338 ± 5 ms
18.2 - training | batch=10, size=1024x300: 1243 ± 20 ms

19/19. GNMT-Translation

19.1 - inference | batch=1, size=1x20: 134 ± 3 ms