why add detach() ?

[manmaynakhashi]

why do you detach()

Avocodo-pytorch/train.py

Line 130 in b298072

y_df_hat_r, y_df_hat_g, _, _ = mcmbd(y, y_g_hat.detach(), x2.detach(), x1.detach())

and not at

Avocodo-pytorch/train.py

Line 148 in b298072

y_df_hat_r, y_df_hat_g, fmap_f_r, fmap_f_g = mcmbd(y, y_g_hat, x2, x1)

@rishikksh20

[v-nhandt21]

The Collaborative Multi-Band Discriminator in the line 130 is for updating the discriminator only, we take the detach to do copy to new tensor so that the generator don't get gradient.

In the line 148, we will update gradient for the generator