diffusers 0.17.1__py3-none-any.whl → 0.18.2__py3-none-any.whl

diffusers/schedulers/scheduling_lms_discrete.py

@@ -45,7 +45,11 @@ class LMSDiscreteSchedulerOutput(BaseOutput):
 
 
 # Copied from diffusers.schedulers.scheduling_ddpm.betas_for_alpha_bar
-def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
+def betas_for_alpha_bar(
+    num_diffusion_timesteps,
+    max_beta=0.999,
+    alpha_transform_type="cosine",
+):
     """
     Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of
     (1-beta) over time from t = [0,1].
@@ -58,19 +62,30 @@ def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
         num_diffusion_timesteps (`int`): the number of betas to produce.
         max_beta (`float`): the maximum beta to use; use values lower than 1 to
                      prevent singularities.
+        alpha_transform_type (`str`, *optional*, default to `cosine`): the type of noise schedule for alpha_bar.
+                     Choose from `cosine` or `exp`
 
     Returns:
         betas (`np.ndarray`): the betas used by the scheduler to step the model outputs
     """
+    if alpha_transform_type == "cosine":
 
-    def alpha_bar(time_step):
-        return math.cos((time_step + 0.008) / 1.008 * math.pi / 2) ** 2
+        def alpha_bar_fn(t):
+            return math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2
+
+    elif alpha_transform_type == "exp":
+
+        def alpha_bar_fn(t):
+            return math.exp(t * -12.0)
+
+    else:
+        raise ValueError(f"Unsupported alpha_tranform_type: {alpha_transform_type}")
 
     betas = []
     for i in range(num_diffusion_timesteps):
         t1 = i / num_diffusion_timesteps
         t2 = (i + 1) / num_diffusion_timesteps
-        betas.append(min(1 - alpha_bar(t2) / alpha_bar(t1), max_beta))
+        betas.append(min(1 - alpha_bar_fn(t2) / alpha_bar_fn(t1), max_beta))
     return torch.tensor(betas, dtype=torch.float32)
 
 
@@ -102,6 +117,13 @@ class LMSDiscreteScheduler(SchedulerMixin, ConfigMixin):
             prediction type of the scheduler function, one of `epsilon` (predicting the noise of the diffusion
             process), `sample` (directly predicting the noisy sample`) or `v_prediction` (see section 2.4
             https://imagen.research.google/video/paper.pdf)
+        timestep_spacing (`str`, default `"linspace"`):
+            The way the timesteps should be scaled. Refer to Table 2. of [Common Diffusion Noise Schedules and Sample
+            Steps are Flawed](https://arxiv.org/abs/2305.08891) for more information.
+        steps_offset (`int`, default `0`):
+            an offset added to the inference steps. You can use a combination of `offset=1` and
+            `set_alpha_to_one=False`, to make the last step use step 0 for the previous alpha product, as done in
+            stable diffusion.
     """
 
     _compatibles = [e.name for e in KarrasDiffusionSchedulers]
@@ -117,6 +139,8 @@ class LMSDiscreteScheduler(SchedulerMixin, ConfigMixin):
         trained_betas: Optional[Union[np.ndarray, List[float]]] = None,
         use_karras_sigmas: Optional[bool] = False,
         prediction_type: str = "epsilon",
+        timestep_spacing: str = "linspace",
+        steps_offset: int = 0,
     ):
         if trained_betas is not None:
             self.betas = torch.tensor(trained_betas, dtype=torch.float32)
@@ -140,9 +164,6 @@ class LMSDiscreteScheduler(SchedulerMixin, ConfigMixin):
         sigmas = np.concatenate([sigmas[::-1], [0.0]]).astype(np.float32)
         self.sigmas = torch.from_numpy(sigmas)
 
-        # standard deviation of the initial noise distribution
-        self.init_noise_sigma = self.sigmas.max()
-
         # setable values
         self.num_inference_steps = None
         self.use_karras_sigmas = use_karras_sigmas
@@ -150,6 +171,14 @@ class LMSDiscreteScheduler(SchedulerMixin, ConfigMixin):
         self.derivatives = []
         self.is_scale_input_called = False
 
+    @property
+    def init_noise_sigma(self):
+        # standard deviation of the initial noise distribution
+        if self.config.timestep_spacing in ["linspace", "trailing"]:
+            return self.sigmas.max()
+
+        return (self.sigmas.max() ** 2 + 1) ** 0.5
+
     def scale_model_input(
         self, sample: torch.FloatTensor, timestep: Union[float, torch.FloatTensor]
     ) -> torch.FloatTensor:
@@ -205,7 +234,27 @@ class LMSDiscreteScheduler(SchedulerMixin, ConfigMixin):
         """
         self.num_inference_steps = num_inference_steps
 
-        timesteps = np.linspace(0, self.config.num_train_timesteps - 1, num_inference_steps, dtype=float)[::-1].copy()
+        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
+        if self.config.timestep_spacing == "linspace":
+            timesteps = np.linspace(0, self.config.num_train_timesteps - 1, num_inference_steps, dtype=float)[
+                ::-1
+            ].copy()
+        elif self.config.timestep_spacing == "leading":
+            step_ratio = self.config.num_train_timesteps // self.num_inference_steps
+            # creates integer timesteps by multiplying by ratio
+            # casting to int to avoid issues when num_inference_step is power of 3
+            timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()[::-1].copy().astype(float)
+            timesteps += self.config.steps_offset
+        elif self.config.timestep_spacing == "trailing":
+            step_ratio = self.config.num_train_timesteps / self.num_inference_steps
+            # creates integer timesteps by multiplying by ratio
+            # casting to int to avoid issues when num_inference_step is power of 3
+            timesteps = (np.arange(self.config.num_train_timesteps, 0, -step_ratio)).round().copy().astype(float)
+            timesteps -= 1
+        else:
+            raise ValueError(
+                f"{self.config.timestep_spacing} is not supported. Please make sure to choose one of 'linspace', 'leading' or 'trailing'."
+            )
 
         sigmas = np.array(((1 - self.alphas_cumprod) / self.alphas_cumprod) ** 0.5)
         log_sigmas = np.log(sigmas)

diffusers/schedulers/scheduling_pndm.py

@@ -25,7 +25,11 @@ from .scheduling_utils import KarrasDiffusionSchedulers, SchedulerMixin, Schedul
 
 
 # Copied from diffusers.schedulers.scheduling_ddpm.betas_for_alpha_bar
-def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
+def betas_for_alpha_bar(
+    num_diffusion_timesteps,
+    max_beta=0.999,
+    alpha_transform_type="cosine",
+):
     """
     Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of
     (1-beta) over time from t = [0,1].
@@ -38,19 +42,30 @@ def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
         num_diffusion_timesteps (`int`): the number of betas to produce.
         max_beta (`float`): the maximum beta to use; use values lower than 1 to
                      prevent singularities.
+        alpha_transform_type (`str`, *optional*, default to `cosine`): the type of noise schedule for alpha_bar.
+                     Choose from `cosine` or `exp`
 
     Returns:
         betas (`np.ndarray`): the betas used by the scheduler to step the model outputs
     """
+    if alpha_transform_type == "cosine":
 
-    def alpha_bar(time_step):
-        return math.cos((time_step + 0.008) / 1.008 * math.pi / 2) ** 2
+        def alpha_bar_fn(t):
+            return math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2
+
+    elif alpha_transform_type == "exp":
+
+        def alpha_bar_fn(t):
+            return math.exp(t * -12.0)
+
+    else:
+        raise ValueError(f"Unsupported alpha_tranform_type: {alpha_transform_type}")
 
     betas = []
     for i in range(num_diffusion_timesteps):
         t1 = i / num_diffusion_timesteps
         t2 = (i + 1) / num_diffusion_timesteps
-        betas.append(min(1 - alpha_bar(t2) / alpha_bar(t1), max_beta))
+        betas.append(min(1 - alpha_bar_fn(t2) / alpha_bar_fn(t1), max_beta))
     return torch.tensor(betas, dtype=torch.float32)
 
 
@@ -85,11 +100,13 @@ class PNDMScheduler(SchedulerMixin, ConfigMixin):
         prediction_type (`str`, default `epsilon`, optional):
             prediction type of the scheduler function, one of `epsilon` (predicting the noise of the diffusion process)
             or `v_prediction` (see section 2.4 https://imagen.research.google/video/paper.pdf)
+        timestep_spacing (`str`, default `"leading"`):
+            The way the timesteps should be scaled. Refer to Table 2. of [Common Diffusion Noise Schedules and Sample
+            Steps are Flawed](https://arxiv.org/abs/2305.08891) for more information.
         steps_offset (`int`, default `0`):
             an offset added to the inference steps. You can use a combination of `offset=1` and
             `set_alpha_to_one=False`, to make the last step use step 0 for the previous alpha product, as done in
             stable diffusion.
-
     """
 
     _compatibles = [e.name for e in KarrasDiffusionSchedulers]
@@ -106,6 +123,7 @@ class PNDMScheduler(SchedulerMixin, ConfigMixin):
         skip_prk_steps: bool = False,
         set_alpha_to_one: bool = False,
         prediction_type: str = "epsilon",
+        timestep_spacing: str = "leading",
         steps_offset: int = 0,
     ):
         if trained_betas is not None:
@@ -159,11 +177,29 @@ class PNDMScheduler(SchedulerMixin, ConfigMixin):
         """
 
         self.num_inference_steps = num_inference_steps
-        step_ratio = self.config.num_train_timesteps // self.num_inference_steps
-        # creates integer timesteps by multiplying by ratio
-        # casting to int to avoid issues when num_inference_step is power of 3
-        self._timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()
-        self._timesteps += self.config.steps_offset
+        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
+        if self.config.timestep_spacing == "linspace":
+            self._timesteps = (
+                np.linspace(0, self.config.num_train_timesteps - 1, num_inference_steps).round().astype(np.int64)
+            )
+        elif self.config.timestep_spacing == "leading":
+            step_ratio = self.config.num_train_timesteps // self.num_inference_steps
+            # creates integer timesteps by multiplying by ratio
+            # casting to int to avoid issues when num_inference_step is power of 3
+            self._timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()
+            self._timesteps += self.config.steps_offset
+        elif self.config.timestep_spacing == "trailing":
+            step_ratio = self.config.num_train_timesteps / self.num_inference_steps
+            # creates integer timesteps by multiplying by ratio
+            # casting to int to avoid issues when num_inference_step is power of 3
+            self._timesteps = np.round(np.arange(self.config.num_train_timesteps, 0, -step_ratio))[::-1].astype(
+                np.int64
+            )
+            self._timesteps -= 1
+        else:
+            raise ValueError(
+                f"{self.config.timestep_spacing} is not supported. Please make sure to choose one of 'linspace', 'leading' or 'trailing'."
+            )
 
         if self.config.skip_prk_steps:
             # for some models like stable diffusion the prk steps can/should be skipped to

diffusers/schedulers/scheduling_repaint.py

@@ -43,7 +43,11 @@ class RePaintSchedulerOutput(BaseOutput):
 
 
 # Copied from diffusers.schedulers.scheduling_ddpm.betas_for_alpha_bar
-def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
+def betas_for_alpha_bar(
+    num_diffusion_timesteps,
+    max_beta=0.999,
+    alpha_transform_type="cosine",
+):
     """
     Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of
     (1-beta) over time from t = [0,1].
@@ -56,19 +60,30 @@ def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
         num_diffusion_timesteps (`int`): the number of betas to produce.
         max_beta (`float`): the maximum beta to use; use values lower than 1 to
                      prevent singularities.
+        alpha_transform_type (`str`, *optional*, default to `cosine`): the type of noise schedule for alpha_bar.
+                     Choose from `cosine` or `exp`
 
     Returns:
         betas (`np.ndarray`): the betas used by the scheduler to step the model outputs
     """
+    if alpha_transform_type == "cosine":
 
-    def alpha_bar(time_step):
-        return math.cos((time_step + 0.008) / 1.008 * math.pi / 2) ** 2
+        def alpha_bar_fn(t):
+            return math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2
+
+    elif alpha_transform_type == "exp":
+
+        def alpha_bar_fn(t):
+            return math.exp(t * -12.0)
+
+    else:
+        raise ValueError(f"Unsupported alpha_tranform_type: {alpha_transform_type}")
 
     betas = []
     for i in range(num_diffusion_timesteps):
         t1 = i / num_diffusion_timesteps
         t2 = (i + 1) / num_diffusion_timesteps
-        betas.append(min(1 - alpha_bar(t2) / alpha_bar(t1), max_beta))
+        betas.append(min(1 - alpha_bar_fn(t2) / alpha_bar_fn(t1), max_beta))
     return torch.tensor(betas, dtype=torch.float32)
 
 

diffusers/schedulers/scheduling_sde_ve.py

@@ -276,7 +276,11 @@ class ScoreSdeVeScheduler(SchedulerMixin, ConfigMixin):
         # Make sure sigmas and timesteps have the same device and dtype as original_samples
         timesteps = timesteps.to(original_samples.device)
         sigmas = self.discrete_sigmas.to(original_samples.device)[timesteps]
-        noise = torch.randn_like(original_samples) * sigmas[:, None, None, None]
+        noise = (
+            noise * sigmas[:, None, None, None]
+            if noise is not None
+            else torch.randn_like(original_samples) * sigmas[:, None, None, None]
+        )
         noisy_samples = noise + original_samples
         return noisy_samples
 

diffusers/schedulers/scheduling_unclip.py

@@ -44,7 +44,11 @@ class UnCLIPSchedulerOutput(BaseOutput):
 
 
 # Copied from diffusers.schedulers.scheduling_ddpm.betas_for_alpha_bar
-def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
+def betas_for_alpha_bar(
+    num_diffusion_timesteps,
+    max_beta=0.999,
+    alpha_transform_type="cosine",
+):
     """
     Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of
     (1-beta) over time from t = [0,1].
@@ -57,19 +61,30 @@ def betas_for_alpha_bar(num_diffusion_timesteps, max_beta=0.999):
         num_diffusion_timesteps (`int`): the number of betas to produce.
         max_beta (`float`): the maximum beta to use; use values lower than 1 to
                      prevent singularities.
+        alpha_transform_type (`str`, *optional*, default to `cosine`): the type of noise schedule for alpha_bar.
+                     Choose from `cosine` or `exp`
 
     Returns:
         betas (`np.ndarray`): the betas used by the scheduler to step the model outputs
     """
+    if alpha_transform_type == "cosine":
 
-    def alpha_bar(time_step):
-        return math.cos((time_step + 0.008) / 1.008 * math.pi / 2) ** 2
+        def alpha_bar_fn(t):
+            return math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2
+
+    elif alpha_transform_type == "exp":
+
+        def alpha_bar_fn(t):
+            return math.exp(t * -12.0)
+
+    else:
+        raise ValueError(f"Unsupported alpha_tranform_type: {alpha_transform_type}")
 
     betas = []
     for i in range(num_diffusion_timesteps):
         t1 = i / num_diffusion_timesteps
         t2 = (i + 1) / num_diffusion_timesteps
-        betas.append(min(1 - alpha_bar(t2) / alpha_bar(t1), max_beta))
+        betas.append(min(1 - alpha_bar_fn(t2) / alpha_bar_fn(t1), max_beta))
     return torch.tensor(betas, dtype=torch.float32)
 
 
@@ -307,3 +322,27 @@ class UnCLIPScheduler(SchedulerMixin, ConfigMixin):
             return (pred_prev_sample,)
 
         return UnCLIPSchedulerOutput(prev_sample=pred_prev_sample, pred_original_sample=pred_original_sample)
+
+    # Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler.add_noise
+    def add_noise(
+        self,
+        original_samples: torch.FloatTensor,
+        noise: torch.FloatTensor,
+        timesteps: torch.IntTensor,
+    ) -> torch.FloatTensor:
+        # Make sure alphas_cumprod and timestep have same device and dtype as original_samples
+        alphas_cumprod = self.alphas_cumprod.to(device=original_samples.device, dtype=original_samples.dtype)
+        timesteps = timesteps.to(original_samples.device)
+
+        sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
+        sqrt_alpha_prod = sqrt_alpha_prod.flatten()
+        while len(sqrt_alpha_prod.shape) < len(original_samples.shape):
+            sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)
+
+        sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
+        sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
+        while len(sqrt_one_minus_alpha_prod.shape) < len(original_samples.shape):
+            sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)
+
+        noisy_samples = sqrt_alpha_prod * original_samples + sqrt_one_minus_alpha_prod * noise
+        return noisy_samples

diffusers/schedulers/scheduling_unipc_multistep.py

@@ -117,6 +117,17 @@ class UniPCMultistepScheduler(SchedulerMixin, ConfigMixin):
             by disable the corrector at the first few steps (e.g., disable_corrector=[0])
         solver_p (`SchedulerMixin`, default `None`):
             can be any other scheduler. If specified, the algorithm will become solver_p + UniC.
+        use_karras_sigmas (`bool`, *optional*, defaults to `False`):
+             This parameter controls whether to use Karras sigmas (Karras et al. (2022) scheme) for step sizes in the
+             noise schedule during the sampling process. If True, the sigmas will be determined according to a sequence
+             of noise levels {σi} as defined in Equation (5) of the paper https://arxiv.org/pdf/2206.00364.pdf.
+        timestep_spacing (`str`, default `"linspace"`):
+            The way the timesteps should be scaled. Refer to Table 2. of [Common Diffusion Noise Schedules and Sample
+            Steps are Flawed](https://arxiv.org/abs/2305.08891) for more information.
+        steps_offset (`int`, default `0`):
+            an offset added to the inference steps. You can use a combination of `offset=1` and
+            `set_alpha_to_one=False`, to make the last step use step 0 for the previous alpha product, as done in
+            stable diffusion.
     """
 
     _compatibles = [e.name for e in KarrasDiffusionSchedulers]
@@ -140,6 +151,9 @@ class UniPCMultistepScheduler(SchedulerMixin, ConfigMixin):
         lower_order_final: bool = True,
         disable_corrector: List[int] = [],
         solver_p: SchedulerMixin = None,
+        use_karras_sigmas: Optional[bool] = False,
+        timestep_spacing: str = "linspace",
+        steps_offset: int = 0,
     ):
         if trained_betas is not None:
             self.betas = torch.tensor(trained_betas, dtype=torch.float32)
@@ -168,7 +182,7 @@ class UniPCMultistepScheduler(SchedulerMixin, ConfigMixin):
 
         if solver_type not in ["bh1", "bh2"]:
             if solver_type in ["midpoint", "heun", "logrho"]:
-                self.register_to_config(solver_type="bh1")
+                self.register_to_config(solver_type="bh2")
             else:
                 raise NotImplementedError(f"{solver_type} does is not implemented for {self.__class__}")
 
@@ -194,12 +208,39 @@ class UniPCMultistepScheduler(SchedulerMixin, ConfigMixin):
             device (`str` or `torch.device`, optional):
                 the device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
         """
-        timesteps = (
-            np.linspace(0, self.config.num_train_timesteps - 1, num_inference_steps + 1)
-            .round()[::-1][:-1]
-            .copy()
-            .astype(np.int64)
-        )
+        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
+        if self.config.timestep_spacing == "linspace":
+            timesteps = (
+                np.linspace(0, self.config.num_train_timesteps - 1, num_inference_steps + 1)
+                .round()[::-1][:-1]
+                .copy()
+                .astype(np.int64)
+            )
+        elif self.config.timestep_spacing == "leading":
+            step_ratio = self.config.num_train_timesteps // (num_inference_steps + 1)
+            # creates integer timesteps by multiplying by ratio
+            # casting to int to avoid issues when num_inference_step is power of 3
+            timesteps = (np.arange(0, num_inference_steps + 1) * step_ratio).round()[::-1][:-1].copy().astype(np.int64)
+            timesteps += self.config.steps_offset
+        elif self.config.timestep_spacing == "trailing":
+            step_ratio = self.config.num_train_timesteps / num_inference_steps
+            # creates integer timesteps by multiplying by ratio
+            # casting to int to avoid issues when num_inference_step is power of 3
+            timesteps = np.arange(self.config.num_train_timesteps, 0, -step_ratio).round().copy().astype(np.int64)
+            timesteps -= 1
+        else:
+            raise ValueError(
+                f"{self.config.timestep_spacing} is not supported. Please make sure to choose one of 'linspace', 'leading' or 'trailing'."
+            )
+
+        sigmas = np.array(((1 - self.alphas_cumprod) / self.alphas_cumprod) ** 0.5)
+        if self.config.use_karras_sigmas:
+            log_sigmas = np.log(sigmas)
+            sigmas = self._convert_to_karras(in_sigmas=sigmas, num_inference_steps=num_inference_steps)
+            timesteps = np.array([self._sigma_to_t(sigma, log_sigmas) for sigma in sigmas]).round()
+            timesteps = np.flip(timesteps).copy().astype(np.int64)
+
+        self.sigmas = torch.from_numpy(sigmas)
 
         # when num_inference_steps == num_train_timesteps, we can end up with
         # duplicates in timesteps.

diffusers/training_utils.py

@@ -1,6 +1,6 @@
 import contextlib
 import copy
-from random import random
+import random
 from typing import Any, Dict, Iterable, Optional, Union
 
 import numpy as np

diffusers/utils/__init__.py

@@ -58,6 +58,7 @@ from .import_utils import (
     is_flax_available,
     is_ftfy_available,
     is_inflect_available,
+    is_invisible_watermark_available,
     is_k_diffusion_available,
     is_k_diffusion_version,
     is_librosa_available,
@@ -103,7 +104,7 @@ if is_torch_available():
     )
     from .torch_utils import maybe_allow_in_graph
 
-from .testing_utils import export_to_video
+from .testing_utils import export_to_gif, export_to_video
 
 
 logger = get_logger(__name__)

diffusers/utils/dummy_pt_objects.py

@@ -210,6 +210,21 @@ class AudioPipelineOutput(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class ConsistencyModelPipeline(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class DanceDiffusionPipeline(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -390,6 +405,21 @@ class ScoreSdeVePipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class CMStochasticIterativeScheduler(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class DDIMInverseScheduler(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -405,6 +435,21 @@ class DDIMInverseScheduler(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class DDIMParallelScheduler(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class DDIMScheduler(metaclass=DummyObject):
     _backends = ["torch"]
 
@@ -420,6 +465,21 @@ class DDIMScheduler(metaclass=DummyObject):
         requires_backends(cls, ["torch"])
 
 
+class DDPMParallelScheduler(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch"])
+
+
 class DDPMScheduler(metaclass=DummyObject):
     _backends = ["torch"]
 

diffusers/utils/dummy_torch_and_transformers_and_invisible_watermark_objects.py

@@ -0,0 +1,32 @@
+# This file is autogenerated by the command `make fix-copies`, do not edit.
+from ..utils import DummyObject, requires_backends
+
+
+class StableDiffusionXLImg2ImgPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers", "invisible_watermark"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers", "invisible_watermark"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers", "invisible_watermark"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers", "invisible_watermark"])
+
+
+class StableDiffusionXLPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers", "invisible_watermark"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers", "invisible_watermark"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers", "invisible_watermark"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers", "invisible_watermark"])