Aryan V S a-r-r-o-w

CogVideoX Official

for file in *; do ffmpeg -y -i "$file" -frames:v 1 "../images/${file%.*}.png"; done

ffmpeg -i a.mp4 -i b.mp4 -i c.mp4 -i d.mp4 -filter_complex hstack=inputs=4 output.mp4

	#VERBOSE=0 torchrun --nproc_per_node 3 self_contained_pp_LOC.py
	import os, random, numpy as np, torch, torch.nn as nn, torch.distributed as dist, torch.nn.functional as F
	from torch.optim import AdamW
	from torch.utils.data import DataLoader, DistributedSampler
	from datasets import load_dataset
	from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer

	STEP, local_rank, world_size, verbose = 0, int(os.environ["LOCAL_RANK"]), int(os.environ["WORLD_SIZE"]), os.environ.get("VERBOSE", "0") == "1"

	def set_all_seed(seed):

	import copy

	import torch
	import torch.nn as nn
	import torch.distributed as dist
	from torch.distributed.tensor.device_mesh import DeviceMesh
	from torch.distributed.tensor import Replicate, Shard, DTensor
	from torch.distributed.tensor.parallel.style import RowwiseParallel, ColwiseParallel, SequenceParallel
	from torch.distributed.tensor.parallel.api import parallelize_module
	from torch._utils import _get_device_module

	import argparse
	import functools
	import json
	import os
	import pathlib
	import psutil
	import time

	import torch
	from diffusers import FluxPipeline

	import torch
	import torch.distributed as dist
	from diffusers import AutoencoderKLWan, WanPipeline
	from diffusers.utils import export_to_video

	from finetrainers._metadata import ParamId, CPInput, CPOutput
	from finetrainers.parallel.ptd import apply_context_parallel
	from finetrainers.models.attention_dispatch import attention_provider, attention_dispatch

	torch.nn.functional.scaled_dot_product_attention = attention_dispatch

	# Reference: https://github.com/arcee-ai/mergekit/blob/488957e8e67c82861ecf63ef761f6bc59122dc74/mergekit/scripts/extract_lora.py

	import argparse

	import torch
	from safetensors.torch import load_file, save_file

	torch.backends.cuda.matmul.allow_tf32 = True
	torch.backends.cudnn.allow_tf32 = True
	torch.backends.cuda.preferred_linalg_library("cusolver")

	import contextlib
	import functools
	import inspect
	import os
	from enum import Enum
	from typing import Any, Callable, Dict, List, Literal, Optional, Tuple, Union

	FINETRAINERS_ATTN_CHECKS = os.environ.get("FINETRAINERS_ATTN_CHECKS", "0").lower() in ("1", "true", "yes")
	FINETRAINERS_ATTN_PROVIDER = os.environ.get("FINETRAINERS_ATTN_PROVIDER", "native").lower()

	import argparse
	import contextlib
	import math
	from typing import List, Optional, Tuple

	import numpy as np
	import torch
	import torch.nn as nn
	import torch._inductor.config
	import torch._higher_order_ops.auto_functionalize as af