nodes/aigc-apps-VideoX-Fun
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Rename && Update Readme

Browse Source
This commit is contained in:
bubbliiiing
2024-09-10 11:22:35 +08:00
parent d29d193177
commit ef55bd3bbe
15 changed files with 107 additions and 132 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/train.py
+4 -4
View File
@@ -72,8 +72,8 @@ from cogvideox.data.dataset_image_video import (ImageVideoDataset,
ImageVideoSampler,
get_random_mask)
from cogvideox.models.transformer3d import CogVideoXTransformer3DModel
from cogvideox.pipeline.pipeline_cogvideox import CogVideoX_FUN_Pipeline
from cogvideox.pipeline.pipeline_cogvideox_inpaint import CogVideoX_FUN_Pipeline_Inpaint
from cogvideox.pipeline.pipeline_cogvideox import CogVideoX_Fun_Pipeline
from cogvideox.pipeline.pipeline_cogvideox_inpaint import CogVideoX_Fun_Pipeline_Inpaint
from cogvideox.utils.utils import get_image_to_video_latent, save_videos_grid
if is_wandb_available():
@@ -163,7 +163,7 @@ def log_validation(vae, text_encoder, tokenizer, transformer3d, args, accelerato
transformer3d_val.load_state_dict(accelerator.unwrap_model(transformer3d).state_dict())
if args.train_mode != "normal":
pipeline = CogVideoX_FUN_Pipeline_Inpaint.from_pretrained(
pipeline = CogVideoX_Fun_Pipeline_Inpaint.from_pretrained(
args.pretrained_model_name_or_path,
vae=accelerator.unwrap_model(vae).to(weight_dtype),
text_encoder=accelerator.unwrap_model(text_encoder),
@@ -172,7 +172,7 @@ def log_validation(vae, text_encoder, tokenizer, transformer3d, args, accelerato
torch_dtype=weight_dtype
)
else:
pipeline = CogVideoX_FUN_Pipeline.from_pretrained(
pipeline = CogVideoX_Fun_Pipeline.from_pretrained(
args.pretrained_model_name_or_path,
vae=accelerator.unwrap_model(vae).to(weight_dtype),
text_encoder=accelerator.unwrap_model(text_encoder),
scripts/train_lora.py
+4 -4
View File
@@ -70,8 +70,8 @@ from cogvideox.data.bucket_sampler import (ASPECT_RATIO_512,
AspectRatioBatchImageVideoSampler,
AspectRatioBatchSampler,
RandomSampler, get_closest_ratio)
from cogvideox.pipeline.pipeline_cogvideox import CogVideoX_FUN_Pipeline
from cogvideox.pipeline.pipeline_cogvideox_inpaint import CogVideoX_FUN_Pipeline_Inpaint
from cogvideox.pipeline.pipeline_cogvideox import CogVideoX_Fun_Pipeline
from cogvideox.pipeline.pipeline_cogvideox_inpaint import CogVideoX_Fun_Pipeline_Inpaint
from cogvideox.data.dataset_image import CC15M
from cogvideox.data.dataset_image_video import (ImageVideoDataset,
ImageVideoSampler,
@@ -168,7 +168,7 @@ def log_validation(vae, text_encoder, tokenizer, transformer3d, network, args, a
transformer3d_val.load_state_dict(accelerator.unwrap_model(transformer3d).state_dict())
if args.train_mode != "normal":
pipeline = CogVideoX_FUN_Pipeline_Inpaint.from_pretrained(
pipeline = CogVideoX_Fun_Pipeline_Inpaint.from_pretrained(
args.pretrained_model_name_or_path,
vae=accelerator.unwrap_model(vae).to(weight_dtype),
text_encoder=accelerator.unwrap_model(text_encoder),
@@ -177,7 +177,7 @@ def log_validation(vae, text_encoder, tokenizer, transformer3d, network, args, a
torch_dtype=weight_dtype,
)
else:
pipeline = CogVideoX_FUN_Pipeline.from_pretrained(
pipeline = CogVideoX_Fun_Pipeline.from_pretrained(
args.pretrained_model_name_or_path,
vae=accelerator.unwrap_model(vae).to(weight_dtype),
text_encoder=accelerator.unwrap_model(text_encoder),