Upload folder using huggingface_hub
Browse files- main/rerender_a_video.py +11 -1
main/rerender_a_video.py
CHANGED
|
@@ -30,10 +30,17 @@ from diffusers.pipelines.controlnet.multicontrolnet import MultiControlNetModel
|
|
| 30 |
from diffusers.pipelines.controlnet.pipeline_controlnet_img2img import StableDiffusionControlNetImg2ImgPipeline
|
| 31 |
from diffusers.pipelines.stable_diffusion.safety_checker import StableDiffusionSafetyChecker
|
| 32 |
from diffusers.schedulers import KarrasDiffusionSchedulers
|
| 33 |
-
from diffusers.utils import BaseOutput, deprecate, logging
|
| 34 |
from diffusers.utils.torch_utils import is_compiled_module, randn_tensor
|
| 35 |
|
| 36 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
|
| 38 |
|
| 39 |
|
|
@@ -1100,6 +1107,9 @@ class RerenderAVideoPipeline(StableDiffusionControlNetImg2ImgPipeline):
|
|
| 1100 |
if callback is not None and i % callback_steps == 0:
|
| 1101 |
callback(i, t, latents)
|
| 1102 |
|
|
|
|
|
|
|
|
|
|
| 1103 |
return latents
|
| 1104 |
|
| 1105 |
if mask_start_t <= mask_end_t:
|
|
|
|
| 30 |
from diffusers.pipelines.controlnet.pipeline_controlnet_img2img import StableDiffusionControlNetImg2ImgPipeline
|
| 31 |
from diffusers.pipelines.stable_diffusion.safety_checker import StableDiffusionSafetyChecker
|
| 32 |
from diffusers.schedulers import KarrasDiffusionSchedulers
|
| 33 |
+
from diffusers.utils import BaseOutput, deprecate, is_torch_xla_available, logging
|
| 34 |
from diffusers.utils.torch_utils import is_compiled_module, randn_tensor
|
| 35 |
|
| 36 |
|
| 37 |
+
if is_torch_xla_available():
|
| 38 |
+
import torch_xla.core.xla_model as xm
|
| 39 |
+
|
| 40 |
+
XLA_AVAILABLE = True
|
| 41 |
+
else:
|
| 42 |
+
XLA_AVAILABLE = False
|
| 43 |
+
|
| 44 |
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
|
| 45 |
|
| 46 |
|
|
|
|
| 1107 |
if callback is not None and i % callback_steps == 0:
|
| 1108 |
callback(i, t, latents)
|
| 1109 |
|
| 1110 |
+
if XLA_AVAILABLE:
|
| 1111 |
+
xm.mark_step()
|
| 1112 |
+
|
| 1113 |
return latents
|
| 1114 |
|
| 1115 |
if mask_start_t <= mask_end_t:
|