refactor

pyproject.toml

@@ -1,20 +0,0 @@
-[tool.poetry]
-name = "xora-core"
-version = "0.1.0"
-description = ""
-authors = ["Your Name <[email protected]>"]
-readme = "README.md"
-
-[tool.poetry.dependencies]
-python = "^3.10"
-torch = "2.3.0"
-diffusers = "0.28.2"
-einops = "0.6.1"
-transformers = "4.44.2"
-sentencepiece = "^0.2.0"
-accelerate = "^0.34.2"
-
-
-[build-system]
-requires = ["poetry-core"]
-build-backend = "poetry.core.masonry.api"

requirements.txt

@@ -0,0 +1,48 @@
+accelerate==0.34.2 ; python_version >= "3.10" and python_version < "4.0"
+certifi==2024.8.30 ; python_version >= "3.10" and python_version < "4.0"
+charset-normalizer==3.3.2 ; python_version >= "3.10" and python_version < "4.0"
+colorama==0.4.6 ; python_version >= "3.10" and python_version < "4.0" and platform_system == "Windows"
+diffusers==0.28.2 ; python_version >= "3.10" and python_version < "4.0"
+einops==0.6.1 ; python_version >= "3.10" and python_version < "4.0"
+filelock==3.16.1 ; python_version >= "3.10" and python_version < "4.0"
+fsspec==2024.9.0 ; python_version >= "3.10" and python_version < "4.0"
+huggingface-hub==0.25.1 ; python_version >= "3.10" and python_version < "4.0"
+idna==3.10 ; python_version >= "3.10" and python_version < "4.0"
+importlib-metadata==8.5.0 ; python_version >= "3.10" and python_version < "4.0"
+intel-openmp==2021.4.0 ; python_version >= "3.10" and python_version < "4.0" and platform_system == "Windows"
+jinja2==3.1.4 ; python_version >= "3.10" and python_version < "4.0"
+markupsafe==2.1.5 ; python_version >= "3.10" and python_version < "4.0"
+mkl==2021.4.0 ; python_version >= "3.10" and python_version < "4.0" and platform_system == "Windows"
+mpmath==1.3.0 ; python_version >= "3.10" and python_version < "4.0"
+networkx==3.3 ; python_version >= "3.10" and python_version < "4.0"
+numpy==2.1.1 ; python_version >= "3.10" and python_version < "4.0"
+nvidia-cublas-cu12==12.1.3.1 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-cuda-cupti-cu12==12.1.105 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-cuda-nvrtc-cu12==12.1.105 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-cuda-runtime-cu12==12.1.105 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-cudnn-cu12==8.9.2.26 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-cufft-cu12==11.0.2.54 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-curand-cu12==10.3.2.106 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-cusolver-cu12==11.4.5.107 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-cusparse-cu12==12.1.0.106 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-nccl-cu12==2.20.5 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-nvjitlink-cu12==12.6.68 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+nvidia-nvtx-cu12==12.1.105 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version >= "3.10" and python_version < "4.0"
+packaging==24.1 ; python_version >= "3.10" and python_version < "4.0"
+pillow==10.4.0 ; python_version >= "3.10" and python_version < "4.0"
+psutil==6.0.0 ; python_version >= "3.10" and python_version < "4.0"
+pyyaml==6.0.2 ; python_version >= "3.10" and python_version < "4.0"
+regex==2024.9.11 ; python_version >= "3.10" and python_version < "4.0"
+requests==2.32.3 ; python_version >= "3.10" and python_version < "4.0"
+safetensors==0.4.5 ; python_version >= "3.10" and python_version < "4.0"
+sentencepiece==0.2.0 ; python_version >= "3.10" and python_version < "4.0"
+sympy==1.13.3 ; python_version >= "3.10" and python_version < "4.0"
+tbb==2021.13.1 ; python_version >= "3.10" and python_version < "4.0" and platform_system == "Windows"
+tokenizers==0.19.1 ; python_version >= "3.10" and python_version < "4.0"
+torch==2.3.0 ; python_version >= "3.10" and python_version < "4.0"
+tqdm==4.66.5 ; python_version >= "3.10" and python_version < "4.0"
+transformers==4.44.2 ; python_version >= "3.10" and python_version < "4.0"
+triton==2.3.0 ; platform_system == "Linux" and platform_machine == "x86_64" and python_version < "3.12" and python_version >= "3.10"
+typing-extensions==4.12.2 ; python_version >= "3.10" and python_version < "4.0"
+urllib3==2.2.3 ; python_version >= "3.10" and python_version < "4.0"
+zipp==3.20.2 ; python_version >= "3.10" and python_version < "4.0"

setup.py

@@ -0,0 +1,23 @@
+from setuptools import setup, find_packages
+def parse_requirements(filename):
+    """Load requirements from a pip requirements file."""
+    with open(filename, 'r') as file:
+        return file.read().splitlines()
+
+
+setup(
+    name="xora-core",  # The name of your package
+    version="0.1.0",  # Version number of the package
+    description="A package for Xora model inferece",  # Short description
+    author="YSapir Weissbuch",  # Your name
+    author_email="[email protected]",  # Your email
+    url="https://github.com/LightricksResearch/xora-core",  # URL for the project (GitHub, etc.)
+    packages=find_packages(),  # Automatically find all packages inside `xora`
+    install_requires=parse_requirements('requirements.txt'),  # Install dependencies from requirements.txt
+    classifiers=[
+        'Programming Language :: Python :: 3',
+        'License :: OSI Approved :: MIT License',
+        'Operating System :: OS Independent',
+    ],
+    python_requires='>=3.10',  # Specify Python version compatibility
+)

xora/__init__.py

@@ -0,0 +1 @@
+from .pipelines import *

eval.py → xora/examples/text_to_video.py

@@ -1,9 +1,9 @@
 import torch
-from vae.autoencoders.causal_video_autoencoder import CausalVideoAutoencoder
-from transformer.transformer3d import Transformer3DModel
-from patchify.symmetric import SymmetricPatchifier
-from scheduler.rf import RectifiedFlowScheduler
-from pipeline.pipeline_video_pixart_alpha import VideoPixArtAlphaPipeline
+from xora.models.autoencoders.causal_video_autoencoder import CausalVideoAutoencoder
+from xora.models.transformers.transformer3d import Transformer3DModel
+from xora.models.transformers.symmetric_patchifier import SymmetricPatchifier
+from xora.schedulers.rf import RectifiedFlowScheduler
+from xora.pipelines.pipeline_video_pixart_alpha import VideoPixArtAlphaPipeline
 from pathlib import Path
 from transformers import T5EncoderModel
 

{vae → xora/models}/autoencoders/causal_video_autoencoder.py

@@ -9,9 +9,9 @@ import numpy as np
 from einops import rearrange
 from torch import nn
 
-from vae.layers.conv_nd_factory import make_conv_nd, make_linear_nd
-from vae.layers.pixel_norm import PixelNorm
-from vae.vae import AutoencoderKLWrapper
+from xora.models.autoencoders.conv_nd_factory import make_conv_nd, make_linear_nd
+from xora.models.autoencoders.pixel_norm import PixelNorm
+from xora.models.autoencoders.vae import AutoencoderKLWrapper
 
 
 class CausalVideoAutoencoder(AutoencoderKLWrapper):

{vae/layers → xora/models/autoencoders}/conv_nd_factory.py

@@ -2,8 +2,8 @@ from typing import Tuple, Union
 
 import torch
 
-from vae.layers.dual_conv3d import DualConv3d
-from vae.layers.causal_conv3d import CausalConv3d
+from xora.models.autoencoders.dual_conv3d import DualConv3d
+from xora.models.autoencoders.causal_conv3d import CausalConv3d
 
 
 def make_conv_nd(

{vae → xora/models/autoencoders}/vae.py

@@ -6,7 +6,7 @@ import torch.nn as nn
 from diffusers import ConfigMixin, ModelMixin
 from diffusers.models.autoencoders.vae import DecoderOutput, DiagonalGaussianDistribution
 from diffusers.models.modeling_outputs import AutoencoderKLOutput
-from vae.layers.conv_nd_factory import make_conv_nd
+from xora.models.autoencoders.conv_nd_factory import make_conv_nd
 
 
 class AutoencoderKLWrapper(ModelMixin, ConfigMixin):

{vae → xora/models/autoencoders}/vae_encode.py

@@ -6,7 +6,7 @@ from torch import Tensor
 from torch.nn import functional
 
 
-from vae.autoencoders.causal_video_autoencoder import CausalVideoAutoencoder
+from xora.models.autoencoders.causal_video_autoencoder import CausalVideoAutoencoder
 
 class Downsample3D(nn.Module):
     def __init__(self, dims, in_channels: int, out_channels: int, kernel_size: int = 3, padding: int = 1):

patchify/symmetric.py → xora/models/transformers/symmetric_patchifier.py

@@ -6,7 +6,7 @@ from diffusers.configuration_utils import ConfigMixin
 from einops import rearrange
 from torch import Tensor
 
-from utils.torch_utils import append_dims
+from xora.utils.torch_utils import append_dims
 
 
 class Patchifier(ConfigMixin, ABC):

{transformer → xora/models/transformers}/transformer3d.py

@@ -11,7 +11,7 @@ from diffusers.models.normalization import AdaLayerNormSingle
 from diffusers.utils import BaseOutput, is_torch_version
 from torch import nn
 
-from transformer.attention import BasicTransformerBlock
+from xora.models.transformers.attention import BasicTransformerBlock
 
 
 @dataclass

{pipeline → xora/pipelines}/pipeline_video_pixart_alpha.py

@@ -27,11 +27,11 @@ from diffusers.utils.torch_utils import randn_tensor
 from einops import rearrange
 from transformers import T5EncoderModel, T5Tokenizer
 
-from transformer.transformer3d import Transformer3DModel
-from patchify.symmetric import Patchifier
-from vae.vae_encode import get_vae_size_scale_factor, vae_decode, vae_encode
-from vae.autoencoders.causal_video_autoencoder import CausalVideoAutoencoder
-from scheduler.rf import TimestepShifter
+from xora.models.transformers.transformer3d import Transformer3DModel
+from xora.models.transformers.symmetric_patchifier import Patchifier
+from xora.models.autoencoders.vae_encode import get_vae_size_scale_factor, vae_decode, vae_encode
+from xora.models.autoencoders.causal_video_autoencoder import CausalVideoAutoencoder
+from xora.schedulers.rf import TimestepShifter
 
 logger = logging.get_logger(__name__)  # pylint: disable=invalid-name
 

{scheduler → xora/schedulers}/rf.py

@@ -9,7 +9,7 @@ from diffusers.schedulers.scheduling_utils import SchedulerMixin
 from diffusers.utils import BaseOutput
 from torch import Tensor
 
-from utils.torch_utils import append_dims
+from xora.utils.torch_utils import append_dims
 
 
 def simple_diffusion_resolution_dependent_timestep_shift(