Back to xformers

5 files changed, 14 insertions, 12 deletions

diff --git a/environment.yaml b/environment.yaml
index 8010c09..325644f 100644
--- a/environment.yaml
+++ b/environment.yaml
@@ -1,17 +1,18 @@
 name: ldd
 channels:
-  - pytorch-nightly
+  - pytorch
   - nvidia
   - xformers/label/dev
   - defaults
 dependencies:
   - cudatoolkit=11.7
+  - libcufile=1.4.0.31
   - matplotlib=3.6.2
   - numpy=1.23.4
   - pip=22.3.1
   - python=3.10.8
-  - pytorch=2.0.0.dev20230216=*cuda*
-  - torchvision=0.15.0.dev20230216
+  - pytorch=1.13.1=*cuda*
+  - torchvision=0.14.1
   - pip:
       - -e .
       - -e git+https://github.com/huggingface/diffusers#egg=diffusers
@@ -22,3 +23,5 @@ dependencies:
       - setuptools==65.6.3
       - test-tube>=0.7.5
       - transformers==4.26.1
+      - triton==2.0.0a2
+      - xformers==0.0.17.dev451
diff --git a/infer.py b/infer.py
index 13219f8..80bd208 100644
--- a/infer.py
+++ b/infer.py
@@ -245,8 +245,8 @@ def create_pipeline(model, dtype):
         tokenizer=tokenizer,
         scheduler=scheduler,
     )
-    # pipeline.enable_xformers_memory_efficient_attention()
-    pipeline.unet = torch.compile(pipeline.unet)
+    pipeline.enable_xformers_memory_efficient_attention()
+    # pipeline.unet = torch.compile(pipeline.unet)
     pipeline.enable_vae_slicing()
     pipeline.to("cuda")
 
diff --git a/train_dreambooth.py b/train_dreambooth.py
index 85b756c..5a7911c 100644
--- a/train_dreambooth.py
+++ b/train_dreambooth.py
@@ -464,8 +464,8 @@ def main():
     tokenizer.set_dropout(args.vector_dropout)
 
     vae.enable_slicing()
-    # vae.set_use_memory_efficient_attention_xformers(True)
-    # unet.enable_xformers_memory_efficient_attention()
+    vae.set_use_memory_efficient_attention_xformers(True)
+    unet.enable_xformers_memory_efficient_attention()
 
     if args.gradient_checkpointing:
         unet.enable_gradient_checkpointing()
diff --git a/train_lora.py b/train_lora.py
index 8a06ae8..330bcd6 100644
--- a/train_lora.py
+++ b/train_lora.py
@@ -421,8 +421,8 @@ def main():
         args.pretrained_model_name_or_path)
 
     vae.enable_slicing()
-    # vae.set_use_memory_efficient_attention_xformers(True)
-    # unet.enable_xformers_memory_efficient_attention()
+    vae.set_use_memory_efficient_attention_xformers(True)
+    unet.enable_xformers_memory_efficient_attention()
 
     unet.to(accelerator.device, dtype=weight_dtype)
     text_encoder.to(accelerator.device, dtype=weight_dtype)
diff --git a/train_ti.py b/train_ti.py
index 7d10317..3aa1027 100644
--- a/train_ti.py
+++ b/train_ti.py
@@ -538,9 +538,8 @@ def main():
     tokenizer.set_dropout(args.vector_dropout)
 
     vae.enable_slicing()
-    # vae.set_use_memory_efficient_attention_xformers(True)
-    # unet.enable_xformers_memory_efficient_attention()
-
+    vae.set_use_memory_efficient_attention_xformers(True)
+    unet.enable_xformers_memory_efficient_attention()
     # unet = torch.compile(unet)
 
     if args.gradient_checkpointing: