BIG-PIE-MILK-COW

text-image dataset

7

Is there any text-image dataset that is not so large(containing less than one hundred thousand pairs)?

loss value

18

Does anyone have trained text-conditional model? What about the loss value? I have trained the model on laion-art dataset, and the loss value finally decrease to around 0.1. Is it...

Code for conditional training on dataset which has images and text

8

Is there any code for conditional training on dataset which has images and text?

Is there sometime that sigmaloss may be inf?

1

As defined in SigmaLoss, sigmaloss = -torch.log(weights) * torch.exp(-(z_vals - depths[:, None]) ** 2 / (2 * 1)) * dists, and weights = alpha * torch.cumprod(torch.cat([torch.ones((alpha.shape[0], 1)).to(alpha.device), 1. - alpha...

When I preview at the workbench, the cloth's texture is well, however, if I used eevee to render, the cloth loses it's texture in the result

I have tested the rendering speed and find that it takes 2.8s to render an image on gpu, is there some ways to speed up rendering.