[Cog] some minor fixes and nits

src/diffusers/pipelines/cogvideo/pipeline_cogvideox.py

@@ -316,18 +316,19 @@ def encode_prompt(
     def prepare_latents(
         self, batch_size, num_channels_latents, num_frames, height, width, dtype, device, generator, latents=None
     ):
+        if isinstance(generator, list) and len(generator) != batch_size:
+            raise ValueError(
+                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
         shape = (
             batch_size,
             (num_frames - 1) // self.vae_scale_factor_temporal + 1,
             num_channels_latents,
             height // self.vae_scale_factor_spatial,
             width // self.vae_scale_factor_spatial,
         )
-        if isinstance(generator, list) and len(generator) != batch_size:
-            raise ValueError(
-                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
-                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
-            )
 
         if latents is None:
             latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
@@ -504,10 +505,10 @@ def __call__(
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                 less than `1`).
-            height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
-                The height in pixels of the generated image. This is set to 1024 by default for the best results.
-            width (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
-                The width in pixels of the generated image. This is set to 1024 by default for the best results.
+            height (`int`, *optional*, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial):
+                The height in pixels of the generated image. This is set to 480 by default for the best results.
+            width (`int`, *optional*, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial):
+                The width in pixels of the generated image. This is set to 720 by default for the best results.
             num_frames (`int`, defaults to `48`):
                 Number of frames to generate. Must be divisible by self.vae_scale_factor_temporal. Generated video will
                 contain 1 extra frame because CogVideoX is conditioned with (num_seconds * fps + 1) frames where
@@ -577,8 +578,6 @@ def __call__(
         if isinstance(callback_on_step_end, (PipelineCallback, MultiPipelineCallbacks)):
             callback_on_step_end_tensor_inputs = callback_on_step_end.tensor_inputs
 
-        height = height or self.transformer.config.sample_size * self.vae_scale_factor_spatial
-        width = width or self.transformer.config.sample_size * self.vae_scale_factor_spatial
         num_videos_per_prompt = 1
 
         # 1. Check inputs. Raise error if not correct

src/diffusers/pipelines/cogvideo/pipeline_cogvideox_image2video.py

@@ -207,6 +207,9 @@ def __init__(
         self.vae_scale_factor_temporal = (
             self.vae.config.temporal_compression_ratio if hasattr(self, "vae") and self.vae is not None else 4
         )
+        self.vae_scaling_factor_image = (
+            self.vae.config.scaling_factor if hasattr(self, "vae") and self.vae is not None else 0.7
+        )
 
         self.video_processor = VideoProcessor(vae_scale_factor=self.vae_scale_factor_spatial)
 
@@ -348,6 +351,12 @@ def prepare_latents(
         generator: Optional[torch.Generator] = None,
         latents: Optional[torch.Tensor] = None,
     ):
+        if isinstance(generator, list) and len(generator) != batch_size:
+            raise ValueError(
+                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
         num_frames = (num_frames - 1) // self.vae_scale_factor_temporal + 1
         shape = (
             batch_size,
@@ -357,12 +366,6 @@ def prepare_latents(
             width // self.vae_scale_factor_spatial,
         )
 
-        if isinstance(generator, list) and len(generator) != batch_size:
-            raise ValueError(
-                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
-                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
-            )
-
         image = image.unsqueeze(2)  # [B, C, F, H, W]
 
         if isinstance(generator, list):
@@ -373,7 +376,7 @@ def prepare_latents(
             image_latents = [retrieve_latents(self.vae.encode(img.unsqueeze(0)), generator) for img in image]
 
         image_latents = torch.cat(image_latents, dim=0).to(dtype).permute(0, 2, 1, 3, 4)  # [B, F, C, H, W]
-        image_latents = self.vae.config.scaling_factor * image_latents
+        image_latents = self.vae_scaling_factor_image * image_latents
 
         padding_shape = (
             batch_size,
@@ -397,7 +400,7 @@ def prepare_latents(
     # Copied from diffusers.pipelines.cogvideo.pipeline_cogvideox.CogVideoXPipeline.decode_latents
     def decode_latents(self, latents: torch.Tensor) -> torch.Tensor:
         latents = latents.permute(0, 2, 1, 3, 4)  # [batch_size, num_channels, num_frames, height, width]
-        latents = 1 / self.vae.config.scaling_factor * latents
+        latents = 1 / self.vae_scaling_factor_image * latents
 
         frames = self.vae.decode(latents).sample
         return frames
@@ -438,7 +441,6 @@ def check_inputs(
         width,
         negative_prompt,
         callback_on_step_end_tensor_inputs,
-        video=None,
         latents=None,
         prompt_embeds=None,
         negative_prompt_embeds=None,
@@ -494,9 +496,6 @@ def check_inputs(
                     f" {negative_prompt_embeds.shape}."
                 )
 
-        if video is not None and latents is not None:
-            raise ValueError("Only one of `video` or `latents` should be provided")
-
     # Copied from diffusers.pipelines.cogvideo.pipeline_cogvideox.CogVideoXPipeline.fuse_qkv_projections
     def fuse_qkv_projections(self) -> None:
         r"""Enables fused QKV projections."""
@@ -584,18 +583,18 @@ def __call__(
 
         Args:
             image (`PipelineImageInput`):
-                The input video to condition the generation on. Must be an image, a list of images or a `torch.Tensor`.
+                The input image to condition the generation on. Must be an image, a list of images or a `torch.Tensor`.
             prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
                 instead.
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                 less than `1`).
-            height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
-                The height in pixels of the generated image. This is set to 1024 by default for the best results.
-            width (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
-                The width in pixels of the generated image. This is set to 1024 by default for the best results.
+            height (`int`, *optional*, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial):
+                The height in pixels of the generated image. This is set to 480 by default for the best results.
+            width (`int`, *optional*, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial):
+                The width in pixels of the generated image. This is set to 720 by default for the best results.
             num_frames (`int`, defaults to `48`):
                 Number of frames to generate. Must be divisible by self.vae_scale_factor_temporal. Generated video will
                 contain 1 extra frame because CogVideoX is conditioned with (num_seconds * fps + 1) frames where
@@ -665,20 +664,19 @@ def __call__(
         if isinstance(callback_on_step_end, (PipelineCallback, MultiPipelineCallbacks)):
             callback_on_step_end_tensor_inputs = callback_on_step_end.tensor_inputs
 
-        height = height or self.transformer.config.sample_size * self.vae_scale_factor_spatial
-        width = width or self.transformer.config.sample_size * self.vae_scale_factor_spatial
         num_videos_per_prompt = 1
 
         # 1. Check inputs. Raise error if not correct
         self.check_inputs(
-            image,
-            prompt,
-            height,
-            width,
-            negative_prompt,
-            callback_on_step_end_tensor_inputs,
-            prompt_embeds,
-            negative_prompt_embeds,
+            image=image,
+            prompt=prompt,
+            height=height,
+            width=width,
+            negative_prompt=negative_prompt,
+            callback_on_step_end_tensor_inputs=callback_on_step_end_tensor_inputs,
+            latents=latents,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
         )
         self._guidance_scale = guidance_scale
         self._interrupt = False

src/diffusers/pipelines/cogvideo/pipeline_cogvideox_video2video.py

@@ -206,6 +206,7 @@ def __init__(
         self.register_modules(
             tokenizer=tokenizer, text_encoder=text_encoder, vae=vae, transformer=transformer, scheduler=scheduler
         )
+
         self.vae_scale_factor_spatial = (
             2 ** (len(self.vae.config.block_out_channels) - 1) if hasattr(self, "vae") and self.vae is not None else 8
         )
@@ -353,6 +354,12 @@ def prepare_latents(
         latents: Optional[torch.Tensor] = None,
         timestep: Optional[torch.Tensor] = None,
     ):
+        if isinstance(generator, list) and len(generator) != batch_size:
+            raise ValueError(
+                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
         num_frames = (video.size(2) - 1) // self.vae_scale_factor_temporal + 1 if latents is None else latents.size(1)
 
         shape = (
@@ -363,12 +370,6 @@ def prepare_latents(
             width // self.vae_scale_factor_spatial,
         )
 
-        if isinstance(generator, list) and len(generator) != batch_size:
-            raise ValueError(
-                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
-                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
-            )
-
         if latents is None:
             if isinstance(generator, list):
                 if len(generator) != batch_size:
@@ -586,10 +587,10 @@ def __call__(
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                 less than `1`).
-            height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
-                The height in pixels of the generated image. This is set to 1024 by default for the best results.
-            width (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
-                The width in pixels of the generated image. This is set to 1024 by default for the best results.
+            height (`int`, *optional*, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial):
+                The height in pixels of the generated image. This is set to 480 by default for the best results.
+            width (`int`, *optional*, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial):
+                The width in pixels of the generated image. This is set to 720 by default for the best results.
             num_inference_steps (`int`, *optional*, defaults to 50):
                 The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                 expense of slower inference.
@@ -651,20 +652,20 @@ def __call__(
         if isinstance(callback_on_step_end, (PipelineCallback, MultiPipelineCallbacks)):
             callback_on_step_end_tensor_inputs = callback_on_step_end.tensor_inputs
 
-        height = height or self.transformer.config.sample_size * self.vae_scale_factor_spatial
-        width = width or self.transformer.config.sample_size * self.vae_scale_factor_spatial
         num_videos_per_prompt = 1
 
         # 1. Check inputs. Raise error if not correct
         self.check_inputs(
-            prompt,
-            height,
-            width,
-            strength,
-            negative_prompt,
-            callback_on_step_end_tensor_inputs,
-            prompt_embeds,
-            negative_prompt_embeds,
+            prompt=prompt,
+            height=height,
+            width=width,
+            strength=strength,
+            negative_prompt=negative_prompt,
+            callback_on_step_end_tensor_inputs=callback_on_step_end_tensor_inputs,
+            video=video,
+            latents=latents,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
         )
         self._guidance_scale = guidance_scale
         self._interrupt = False