59 lines
2.5 KiB
Python
Executable File
59 lines
2.5 KiB
Python
Executable File
import importlib.util
|
|
|
|
from transformers import AutoTokenizer, T5EncoderModel, T5Tokenizer
|
|
|
|
from .cogvideox_transformer3d import CogVideoXTransformer3DModel
|
|
from .cogvideox_vae import AutoencoderKLCogVideoX
|
|
from .wan_image_encoder import CLIPModel
|
|
from .wan_text_encoder import WanT5EncoderModel
|
|
from .wan_transformer3d import (Wan2_2Transformer3DModel, WanSelfAttention,
|
|
WanTransformer3DModel)
|
|
from .wan_vae import AutoencoderKLWan, AutoencoderKLWan_
|
|
|
|
# The pai_fuser is an internally developed acceleration package, which can be used on PAI.
|
|
if importlib.util.find_spec("pai_fuser") is not None:
|
|
# The simple_wrapper is used to solve the problem about conflicts between cython and torch.compile
|
|
def simple_wrapper(func):
|
|
def inner(*args, **kwargs):
|
|
return func(*args, **kwargs)
|
|
return inner
|
|
|
|
from ..dist import parallel_magvit_vae
|
|
AutoencoderKLWan_.decode = simple_wrapper(parallel_magvit_vae(0.2, 8)(AutoencoderKLWan_.decode))
|
|
|
|
import torch
|
|
from pai_fuser.core.attention import wan_sparse_attention_wrapper
|
|
|
|
WanSelfAttention.forward = simple_wrapper(wan_sparse_attention_wrapper()(WanSelfAttention.forward))
|
|
print("Import Sparse Attention")
|
|
|
|
WanTransformer3DModel.forward = simple_wrapper(WanTransformer3DModel.forward)
|
|
|
|
import os
|
|
from pai_fuser.core import (cfg_skip_turbo, disable_cfg_skip,
|
|
enable_cfg_skip)
|
|
|
|
WanTransformer3DModel.enable_cfg_skip = enable_cfg_skip()(WanTransformer3DModel.enable_cfg_skip)
|
|
WanTransformer3DModel.disable_cfg_skip = disable_cfg_skip()(WanTransformer3DModel.disable_cfg_skip)
|
|
print("Import CFG Skip Turbo")
|
|
|
|
from pai_fuser.core.rope import ENABLE_KERNEL, fast_rope_apply_qk
|
|
|
|
if ENABLE_KERNEL:
|
|
import types
|
|
from . import wan_transformer3d
|
|
|
|
def deepcopy_function(f):
|
|
return types.FunctionType(f.__code__, f.__globals__, name=f.__name__, argdefs=f.__defaults__,closure=f.__closure__)
|
|
|
|
local_rope_apply_qk = deepcopy_function(wan_transformer3d.rope_apply_qk)
|
|
def adaptive_fast_rope_apply_qk(q, k, grid_sizes, freqs):
|
|
if torch.is_grad_enabled():
|
|
return local_rope_apply_qk(q, k, grid_sizes, freqs)
|
|
else:
|
|
return fast_rope_apply_qk(q, k, grid_sizes, freqs)
|
|
|
|
wan_transformer3d.rope_apply_qk = adaptive_fast_rope_apply_qk
|
|
rope_apply_qk = adaptive_fast_rope_apply_qk
|
|
print("Import PAI Fast rope")
|