close #3304

Link to PR in diffusers repository:
https://github.com/huggingface/diffusers/pull/2691

Imports:
`diffusers.models.cross_attention ->
diffusers.models.attention_processor`

Unions:
`AttnProcessor -> AttentionProcessor`

Classes:
| Old name | New name |
| --- | --- |
| CrossAttention | Attention |
| CrossAttnProcessor | AttnProcessor |
| XFormersCrossAttnProcessor | XFormersAttnProcessor |
| CrossAttnAddedKVProcessor | AttnAddedKVProcessor |
| LoRACrossAttnProcessor | LoRAAttnProcessor |
| LoRAXFormersCrossAttnProcessor | LoRAXFormersAttnProcessor |
| FlaxCrossAttention | FlaxAttention |
| AttendExciteCrossAttnProcessor | AttendExciteAttnProcessor |
| Pix2PixZeroCrossAttnProcessor | Pix2PixZeroAttnProcessor |


Also config values no longer sets as attributes of object:
https://github.com/huggingface/diffusers/pull/2849

.gitignore

@@ -233,5 +233,3 @@ installer/install.sh
 installer/update.bat
 installer/update.sh
 
-# no longer stored in source directory
-models

ldm/invoke/_version.py

@@ -1 +1 @@
-__version__='2.3.5-rc1'
+__version__='2.3.5-rc3'

ldm/invoke/config/model_install_backend.py

@@ -111,7 +111,6 @@ def install_requested_models(
     if len(external_models)>0:
         print("== INSTALLING EXTERNAL MODELS ==")
         for path_url_or_repo in external_models:
-            print(f'DEBUG: path_url_or_repo = {path_url_or_repo}')
             try:
                 model_manager.heuristic_import(
                     path_url_or_repo,

ldm/invoke/generator/diffusers_pipeline.py

@@ -400,8 +400,15 @@ class StableDiffusionGeneratorPipeline(StableDiffusionPipeline):
     @property
     def _submodels(self) -> Sequence[torch.nn.Module]:
         module_names, _, _ = self.extract_init_dict(dict(self.config))
-        values = [getattr(self, name) for name in module_names.keys()]
-        return [m for m in values if isinstance(m, torch.nn.Module)]
+        submodels = []
+        for name in module_names.keys():
+            if hasattr(self, name):
+                value = getattr(self, name)
+            else:
+                value = getattr(self.config, name)
+            if isinstance(value, torch.nn.Module):
+                submodels.append(value)
+        return submodels
 
     def image_from_embeddings(self, latents: torch.Tensor, num_inference_steps: int,
                               conditioning_data: ConditioningData,
@@ -472,7 +479,7 @@ class StableDiffusionGeneratorPipeline(StableDiffusionPipeline):
                                                                 step_count=len(self.scheduler.timesteps)
                                                              ):
 
-            yield PipelineIntermediateState(run_id=run_id, step=-1, timestep=self.scheduler.num_train_timesteps,
+            yield PipelineIntermediateState(run_id=run_id, step=-1, timestep=self.scheduler.config.num_train_timesteps,
                                             latents=latents)
 
             batch_size = latents.shape[0]
@@ -756,7 +763,7 @@ class StableDiffusionGeneratorPipeline(StableDiffusionPipeline):
     @property
     def channels(self) -> int:
         """Compatible with DiffusionWrapper"""
-        return self.unet.in_channels
+        return self.unet.config.in_channels
 
     def decode_latents(self, latents):
         # Explicit call to get the vae loaded, since `decode` isn't the forward method.

ldm/invoke/model_manager.py

@@ -1156,7 +1156,7 @@ class ModelManager(object):
         return self.device.type == "cuda"
 
     def _diffuser_sha256(
-        self, name_or_path: Union[str, Path], chunksize=4096
+        self, name_or_path: Union[str, Path], chunksize=16777216
     ) -> Union[str, bytes]:
         path = None
         if isinstance(name_or_path, Path):

ldm/models/diffusion/cross_attention_control.py

@@ -14,7 +14,6 @@ from torch import nn
 
 from compel.cross_attention_control import Arguments
 from diffusers.models.unet_2d_condition import UNet2DConditionModel
-from diffusers.models.cross_attention import AttnProcessor
 from ldm.invoke.devices import torch_dtype
 
 
@@ -163,7 +162,7 @@ class Context:
 
 class InvokeAICrossAttentionMixin:
     """
-    Enable InvokeAI-flavoured CrossAttention calculation, which does aggressive low-memory slicing and calls
+    Enable InvokeAI-flavoured Attention calculation, which does aggressive low-memory slicing and calls
     through both to an attention_slice_wrangler and a slicing_strategy_getter for custom attention map wrangling
     and dymamic slicing strategy selection.
     """
@@ -178,7 +177,7 @@ class InvokeAICrossAttentionMixin:
         Set custom attention calculator to be called when attention is calculated
         :param wrangler: Callback, with args (module, suggested_attention_slice, dim, offset, slice_size),
         which returns either the suggested_attention_slice or an adjusted equivalent.
-            `module` is the current CrossAttention module for which the callback is being invoked.
+            `module` is the current Attention module for which the callback is being invoked.
             `suggested_attention_slice` is the default-calculated attention slice
             `dim` is -1 if the attenion map has not been sliced, or 0 or 1 for dimension-0 or dimension-1 slicing.
                 If `dim` is >= 0, `offset` and `slice_size` specify the slice start and length.
@@ -326,7 +325,7 @@ def setup_cross_attention_control_attention_processors(unet: UNet2DConditionMode
 
 
 def get_cross_attention_modules(model, which: CrossAttentionType) -> list[tuple[str, InvokeAICrossAttentionMixin]]:
-    from ldm.modules.attention import CrossAttention # avoid circular import
+    from ldm.modules.attention import CrossAttention # avoid circular import # TODO: rename as in diffusers?
     cross_attention_class: type = InvokeAIDiffusersCrossAttention if isinstance(model,UNet2DConditionModel) else CrossAttention
     which_attn = "attn1" if which is CrossAttentionType.SELF else "attn2"
     attention_module_tuples = [(name,module) for name, module in model.named_modules() if
@@ -432,7 +431,7 @@ def get_mem_free_total(device):
 
 
 
-class InvokeAIDiffusersCrossAttention(diffusers.models.attention.CrossAttention, InvokeAICrossAttentionMixin):
+class InvokeAIDiffusersCrossAttention(diffusers.models.attention.Attention, InvokeAICrossAttentionMixin):
 
     def __init__(self, **kwargs):
         super().__init__(**kwargs)
@@ -457,8 +456,8 @@ class InvokeAIDiffusersCrossAttention(diffusers.models.attention.CrossAttention,
 """
 # base implementation
 
-class CrossAttnProcessor:
-    def __call__(self, attn: CrossAttention, hidden_states, encoder_hidden_states=None, attention_mask=None):
+class AttnProcessor:
+    def __call__(self, attn: Attention, hidden_states, encoder_hidden_states=None, attention_mask=None):
         batch_size, sequence_length, _ = hidden_states.shape
         attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length)
 
@@ -487,7 +486,7 @@ from dataclasses import field, dataclass
 
 import torch
 
-from diffusers.models.cross_attention import CrossAttention, CrossAttnProcessor, SlicedAttnProcessor
+from diffusers.models.attention_processor import Attention, AttnProcessor, SlicedAttnProcessor
 
 
 @dataclass
@@ -532,7 +531,7 @@ class SlicedSwapCrossAttnProcesser(SlicedAttnProcessor):
 
     # TODO: dynamically pick slice size based on memory conditions
 
-    def __call__(self, attn: CrossAttention, hidden_states, encoder_hidden_states=None, attention_mask=None,
+    def __call__(self, attn: Attention, hidden_states, encoder_hidden_states=None, attention_mask=None,
                  # kwargs
                  swap_cross_attn_context: SwapCrossAttnContext=None):
 

ldm/models/diffusion/shared_invokeai_diffusion.py

@@ -5,7 +5,6 @@ from typing import Callable, Optional, Union, Any
 
 import numpy as np
 import torch
-
 from diffusers import UNet2DConditionModel
 from typing_extensions import TypeAlias
 

ldm/modules/kohya_lora_manager.py

@@ -9,7 +9,7 @@ from safetensors.torch import load_file
 from torch.utils.hooks import RemovableHandle
 from transformers import CLIPTextModel
 
-from ..invoke.globals import global_lora_models_dir
+from ..invoke.globals import global_lora_models_dir, Globals
 from ..invoke.devices import choose_torch_device
 
 """
@@ -456,16 +456,25 @@ class LoRA:
 
 
 class KohyaLoraManager:
-    lora_path = Path(global_lora_models_dir())
-    vector_length_cache_path = lora_path / '.vectorlength.cache'
     
     def __init__(self, pipe):
+        self.vector_length_cache_path = self.lora_path / '.vectorlength.cache'
         self.unet = pipe.unet
         self.wrapper = LoRAModuleWrapper(pipe.unet, pipe.text_encoder)
         self.text_encoder = pipe.text_encoder
         self.device = torch.device(choose_torch_device())
         self.dtype = pipe.unet.dtype
 
+    @classmethod
+    @property
+    def lora_path(cls)->Path:
+        return Path(global_lora_models_dir())
+
+    @classmethod
+    @property
+    def vector_length_cache_path(cls)->Path:
+        return cls.lora_path / '.vectorlength.cache'        
+
     def load_lora_module(self, name, path_file, multiplier: float = 1.0):
         print(f"   | Found lora {name} at {path_file}")
         if path_file.suffix == ".safetensors":

pyproject.toml

@@ -34,7 +34,7 @@ dependencies = [
   "clip_anytorch",
   "compel~=1.1.0",
   "datasets",
-  "diffusers[torch]==0.14",
+  "diffusers[torch]~=0.15.0",
   "dnspython==2.2.1",
   "einops",
   "eventlet",