Bump to 0.8.0.dev0 (#1131)

* Bump to 0.8.0.dev0

* deprecate int timesteps

* style

setup.py

@@ -210,7 +210,7 @@ install_requires = [
 
 setup(
     name="diffusers",
-    version="0.7.0",  # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots)
+    version="0.8.0.dev0",  # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots)
     description="Diffusers",
     long_description=open("README.md", "r", encoding="utf-8").read(),
     long_description_content_type="text/markdown",

src/diffusers/__init__.py

@@ -9,7 +9,7 @@ from .utils import (
 )
 
 
-__version__ = "0.7.0"
+__version__ = "0.8.0.dev0"
 
 from .configuration_utils import ConfigMixin
 from .onnx_utils import OnnxRuntimeModel

src/diffusers/schedulers/scheduling_euler_ancestral_discrete.py

@@ -19,7 +19,7 @@ import numpy as np
 import torch
 
 from ..configuration_utils import ConfigMixin, register_to_config
-from ..utils import BaseOutput, deprecate, logging
+from ..utils import BaseOutput, logging
 from .scheduling_utils import SchedulerMixin
 
 
@@ -253,19 +253,7 @@ class EulerAncestralDiscreteScheduler(SchedulerMixin, ConfigMixin):
             timesteps = timesteps.to(original_samples.device)
 
         schedule_timesteps = self.timesteps
-
+        step_indices = [(schedule_timesteps == t).nonzero().item() for t in timesteps]
-        if isinstance(timesteps, torch.IntTensor) or isinstance(timesteps, torch.LongTensor):
-            deprecate(
-                "timesteps as indices",
-                "0.8.0",
-                "Passing integer indices  (e.g. from `enumerate(timesteps)`) as timesteps to"
-                " `EulerAncestralDiscreteScheduler.add_noise()` will not be supported in future versions. Make sure to"
-                " pass values from `scheduler.timesteps` as timesteps.",
-                standard_warn=False,
-            )
-            step_indices = timesteps
-        else:
-            step_indices = [(schedule_timesteps == t).nonzero().item() for t in timesteps]
 
         sigma = self.sigmas[step_indices].flatten()
         while len(sigma.shape) < len(original_samples.shape):

src/diffusers/schedulers/scheduling_euler_discrete.py

@@ -19,7 +19,7 @@ import numpy as np
 import torch
 
 from ..configuration_utils import ConfigMixin, register_to_config
-from ..utils import BaseOutput, deprecate, logging
+from ..utils import BaseOutput, logging
 from .scheduling_utils import SchedulerMixin
 
 
@@ -262,19 +262,7 @@ class EulerDiscreteScheduler(SchedulerMixin, ConfigMixin):
             timesteps = timesteps.to(original_samples.device)
 
         schedule_timesteps = self.timesteps
-
+        step_indices = [(schedule_timesteps == t).nonzero().item() for t in timesteps]
-        if isinstance(timesteps, torch.IntTensor) or isinstance(timesteps, torch.LongTensor):
-            deprecate(
-                "timesteps as indices",
-                "0.8.0",
-                "Passing integer indices  (e.g. from `enumerate(timesteps)`) as timesteps to"
-                " `EulerDiscreteScheduler.add_noise()` will not be supported in future versions. Make sure to"
-                " pass values from `scheduler.timesteps` as timesteps.",
-                standard_warn=False,
-            )
-            step_indices = timesteps
-        else:
-            step_indices = [(schedule_timesteps == t).nonzero().item() for t in timesteps]
 
         sigma = self.sigmas[step_indices].flatten()
         while len(sigma.shape) < len(original_samples.shape):

src/diffusers/schedulers/scheduling_lms_discrete.py

@@ -21,7 +21,7 @@ import torch
 from scipy import integrate
 
 from ..configuration_utils import ConfigMixin, register_to_config
-from ..utils import BaseOutput, deprecate
+from ..utils import BaseOutput
 from .scheduling_utils import SchedulerMixin
 
 
@@ -211,22 +211,7 @@ class LMSDiscreteScheduler(SchedulerMixin, ConfigMixin):
 
         if isinstance(timestep, torch.Tensor):
             timestep = timestep.to(self.timesteps.device)
-        if (
+        step_index = (self.timesteps == timestep).nonzero().item()
-            isinstance(timestep, int)
-            or isinstance(timestep, torch.IntTensor)
-            or isinstance(timestep, torch.LongTensor)
-        ):
-            deprecate(
-                "timestep as an index",
-                "0.8.0",
-                "Passing integer indices (e.g. from `enumerate(timesteps)`) as timesteps to"
-                " `LMSDiscreteScheduler.step()` will not be supported in future versions. Make sure to pass"
-                " one of the `scheduler.timesteps` as a timestep.",
-                standard_warn=False,
-            )
-            step_index = timestep
-        else:
-            step_index = (self.timesteps == timestep).nonzero().item()
         sigma = self.sigmas[step_index]
 
         # 1. compute predicted original sample (x_0) from sigma-scaled predicted noise
@@ -269,19 +254,7 @@ class LMSDiscreteScheduler(SchedulerMixin, ConfigMixin):
             timesteps = timesteps.to(original_samples.device)
 
         schedule_timesteps = self.timesteps
-
+        step_indices = [(schedule_timesteps == t).nonzero().item() for t in timesteps]
-        if isinstance(timesteps, torch.IntTensor) or isinstance(timesteps, torch.LongTensor):
-            deprecate(
-                "timesteps as indices",
-                "0.8.0",
-                "Passing integer indices  (e.g. from `enumerate(timesteps)`) as timesteps to"
-                " `LMSDiscreteScheduler.add_noise()` will not be supported in future versions. Make sure to"
-                " pass values from `scheduler.timesteps` as timesteps.",
-                standard_warn=False,
-            )
-            step_indices = timesteps
-        else:
-            step_indices = [(schedule_timesteps == t).nonzero().item() for t in timesteps]
 
         sigma = self.sigmas[step_indices].flatten()
         while len(sigma.shape) < len(original_samples.shape):