Number of iterations for one A100

[postcw]

Due to limited computing resources, I use an A100 to train 300k on FFHQ, batch size = 2. The test effect is not good, how many iterations should be done when there is only one A100?The picture below is the test result：

[e4s2022]

300k iterations. 8 A100 GPU, each with batch size = 2. We use reconstruction as the proxy task during training. So, I think it is better to visualize the reconstruction performance to validate whether the training is normal.

[SeahKwon]

@postcw I have same problem. Did you solve it?

[zsf882299]

hi, @postcw , the weight link of 00000189-checkpoint.pth.tar in the facevid2vid paper is gone,can you send me a copy one？ thank you very much! my email is [email protected]