Add xformers to training scripts (#103)

lora_diffusion/xformers_utils.py

@@ -0,0 +1,70 @@
+import functools
+
+import torch
+from diffusers.models.attention import BasicTransformerBlock
+from diffusers.utils.import_utils import is_xformers_available
+
+from .lora import LoraInjectedLinear
+
+if is_xformers_available():
+    import xformers
+    import xformers.ops
+else:
+    xformers = None
+
+
+@functools.cache
+def test_xformers_backwards(size):
+    @torch.enable_grad()
+    def _grad(size):
+        q = torch.randn((1, 4, size), device="cuda")
+        k = torch.randn((1, 4, size), device="cuda")
+        v = torch.randn((1, 4, size), device="cuda")
+
+        q = q.detach().requires_grad_()
+        k = k.detach().requires_grad_()
+        v = v.detach().requires_grad_()
+
+        out = xformers.ops.memory_efficient_attention(q, k, v)
+        loss = out.sum(2).mean(0).sum()
+
+        return torch.autograd.grad(loss, v)
+
+    try:
+        _grad(size)
+        print(size, "pass")
+        return True
+    except Exception as e:
+        print(size, "fail")
+        return False
+
+
+def set_use_memory_efficient_attention_xformers(
+    module: torch.nn.Module, valid: bool
+) -> None:
+    def fn_test_dim_head(module: torch.nn.Module):
+        if isinstance(module, BasicTransformerBlock):
+            # dim_head isn't stored anywhere, so back-calculate
+            source = module.attn1.to_v
+            if isinstance(source, LoraInjectedLinear):
+                source = source.linear
+
+            dim_head = source.out_features // module.attn1.heads
+
+            result = test_xformers_backwards(dim_head)
+
+            # If dim_head > dim_head_max, turn xformers off
+            if not result:
+                module.set_use_memory_efficient_attention_xformers(False)
+
+        for child in module.children():
+            fn_test_dim_head(child)
+
+    if not is_xformers_available() and valid:
+        print("XFormers is not available. Skipping.")
+        return
+
+    module.set_use_memory_efficient_attention_xformers(valid)
+
+    if valid:
+        fn_test_dim_head(module)

train_lora_dreambooth.py

@@ -36,9 +36,9 @@
     save_lora_weight,
     save_safeloras,
 )
-
-from torch.utils.data import Dataset
+from lora_diffusion.xformers_utils import set_use_memory_efficient_attention_xformers
 from PIL import Image
+from torch.utils.data import Dataset
 from torchvision import transforms
 
 from pathlib import Path
@@ -450,6 +450,9 @@ def parse_args(input_args=None):
         required=False,
         help="Should images be resized to --resolution before training?",
     )
+    parser.add_argument(
+        "--use_xformers", action="store_true", help="Whether or not to use xformers"
+    )
 
     if input_args is not None:
         args = parser.parse_args(input_args)
@@ -615,6 +618,10 @@ def main(args):
             )
             break
 
+    if args.use_xformers:
+        set_use_memory_efficient_attention_xformers(unet, True)
+        set_use_memory_efficient_attention_xformers(vae, True)
+
     if args.gradient_checkpointing:
         unet.enable_gradient_checkpointing()
         if args.train_text_encoder:

train_lora_w_ti.py

@@ -36,9 +36,9 @@
     save_lora_weight,
     extract_lora_ups_down,
 )
-
-from torch.utils.data import Dataset
+from lora_diffusion.xformers_utils import set_use_memory_efficient_attention_xformers
 from PIL import Image
+from torch.utils.data import Dataset
 from torchvision import transforms
 
 from pathlib import Path
@@ -575,6 +575,9 @@ def parse_args(input_args=None):
         required=False,
         help="Should images be resized to --resolution before training?",
     )
+    parser.add_argument(
+        "--use_xformers", action="store_true", help="Whether or not to use xformers"
+    )
 
     if input_args is not None:
         args = parser.parse_args(input_args)
@@ -774,6 +777,10 @@ def main(args):
         print("Before training: text encoder First Layer lora down", _down.weight.data)
         break
 
+    if args.use_xformers:
+        set_use_memory_efficient_attention_xformers(unet, True)
+        set_use_memory_efficient_attention_xformers(vae, True)
+
     if args.gradient_checkpointing:
         unet.enable_gradient_checkpointing()
         if args.train_text_encoder: