画像のアルファチャンネルをlossのマスクとして使用するオプションを追加 (#1223)

* Add alpha_mask parameter and apply masked loss

* Fix type hint in trim_and_resize_if_required function

* Refactor code to use keyword arguments in train_util.py

* Fix alpha mask flipping logic

* Fix alpha mask initialization

* Fix alpha_mask transformation

* Cache alpha_mask

* Update alpha_masks to be on CPU

* Set flipped_alpha_masks to Null if option disabled

* Check if alpha_mask is None

* Set alpha_mask to None if option disabled

* Add description of alpha_mask option to docs

train_textual_inversion_XTI.py

@@ -475,7 +475,9 @@ def train(args):
 
                 loss = train_util.conditional_loss(noise_pred.float(), target.float(), reduction="none", loss_type=args.loss_type, huber_c=huber_c)
                 if args.masked_loss:
-                    loss = apply_masked_loss(loss, batch)
+                    loss = apply_masked_loss(loss, batch["conditioning_images"][:, 0].unsqueeze(1))
+                if "alpha_mask" in batch and batch["alpha_mask"] is not None:
+                    loss = apply_masked_loss(loss, batch["alpha_mask"])
                 loss = loss.mean([1, 2, 3])
 
                 loss_weights = batch["loss_weights"]  # 各sampleごとのweight