undo img2vid changes

examples/cogvideo/README.md

@@ -172,9 +172,6 @@ accelerate launch --gpu_ids $GPU_IDS examples/cogvideo/train_cogvideox_lora.py \
   --report_to wandb
 ```
 
-> [!NOTE]
-> At the time of adding support for CogVideoX-LoRA training, the memory required by the training script, with VAE tiling and LoRA rank 64, is ~52 GB (as tested with the simplest `accelerate config` setting) and ~46 GB (as tested with the simplest `accelerate config` DeepSpeed ZeRO-2 training settings).
-
 To better track our training experiments, we're using the following flags in the command above:
 * `--report_to wandb` will ensure the training runs are tracked on Weights and Biases. To use it, be sure to install `wandb` with `pip install wandb`.
 * `validation_prompt` and `validation_epochs` to allow the script to do a few validation inference runs. This allows us to qualitatively check if the training is progressing as expected.

src/diffusers/pipelines/cogvideo/pipeline_cogvideox_image2video.py

@@ -15,7 +15,7 @@
 
 import inspect
 import math
-from typing import Any, Callable, Dict, List, Optional, Tuple, Union
+from typing import Callable, Dict, List, Optional, Tuple, Union
 
 import PIL
 import torch
@@ -27,7 +27,10 @@ from ...models import AutoencoderKLCogVideoX, CogVideoXTransformer3DModel
 from ...models.embeddings import get_3d_rotary_pos_embed
 from ...pipelines.pipeline_utils import DiffusionPipeline
 from ...schedulers import CogVideoXDDIMScheduler, CogVideoXDPMScheduler
-from ...utils import logging, replace_example_docstring
+from ...utils import (
+    logging,
+    replace_example_docstring,
+)
 from ...utils.torch_utils import randn_tensor
 from ...video_processor import VideoProcessor
 from .pipeline_output import CogVideoXPipelineOutput
@@ -544,10 +547,6 @@ class CogVideoXImageToVideoPipeline(DiffusionPipeline):
     def num_timesteps(self):
         return self._num_timesteps
 
-    @property
-    def attention_kwargs(self):
-        return self._attention_kwargs
-
     @property
     def interrupt(self):
         return self._interrupt
@@ -574,7 +573,6 @@ class CogVideoXImageToVideoPipeline(DiffusionPipeline):
         negative_prompt_embeds: Optional[torch.FloatTensor] = None,
         output_type: str = "pil",
         return_dict: bool = True,
-        attention_kwargs: Optional[Dict[str, Any]] = None,
         callback_on_step_end: Optional[
             Union[Callable[[int, int, Dict], None], PipelineCallback, MultiPipelineCallbacks]
         ] = None,
@@ -638,10 +636,6 @@ class CogVideoXImageToVideoPipeline(DiffusionPipeline):
             return_dict (`bool`, *optional*, defaults to `True`):
                 Whether or not to return a [`~pipelines.stable_diffusion_xl.StableDiffusionXLPipelineOutput`] instead
                 of a plain tuple.
-            attention_kwargs (`dict`, *optional*):
-                A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
-                `self.processor` in
-                [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
             callback_on_step_end (`Callable`, *optional*):
                 A function that calls at the end of each denoising steps during the inference. The function is called
                 with the following arguments: `callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int,
@@ -687,7 +681,6 @@ class CogVideoXImageToVideoPipeline(DiffusionPipeline):
             negative_prompt_embeds,
         )
         self._guidance_scale = guidance_scale
-        self._attention_kwargs = attention_kwargs
         self._interrupt = False
 
         # 2. Default call parameters