Spaces:
Running
on
A10G
Running
on
A10G
File size: 32,448 Bytes
f1069cc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 |
#!/usr/bin/env python
# coding=utf-8
# Copyright 2023 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
import argparse
import logging
import math
import os
import random
from pathlib import Path
import accelerate
import datasets
import numpy as np
import torch
import torch.nn.functional as F
import torch.utils.checkpoint
import transformers
from accelerate import Accelerator
from accelerate.logging import get_logger
from accelerate.utils import ProjectConfiguration, set_seed
from datasets import load_dataset
from huggingface_hub import create_repo, upload_folder
from packaging import version
from torchvision import transforms
from tqdm.auto import tqdm
from transformers import CLIPTextModel, CLIPTokenizer
import diffusers
from diffusers import AutoencoderKL, DDPMScheduler, StableDiffusionPipeline, UNet2DConditionModel
from diffusers.optimization import get_scheduler
from diffusers.training_utils import EMAModel
from diffusers.utils import check_min_version, deprecate
from diffusers.utils.import_utils import is_xformers_available
# Will error if the minimal version of diffusers is not installed. Remove at your own risks.
check_min_version("0.15.0.dev0")
logger = get_logger(__name__, log_level="INFO")
def parse_args():
parser = argparse.ArgumentParser(description="Simple example of a training script.")
parser.add_argument(
"--pretrained_model_name_or_path",
type=str,
default=None,
required=True,
help="Path to pretrained model or model identifier from huggingface.co/models.",
)
parser.add_argument(
"--revision",
type=str,
default=None,
required=False,
help="Revision of pretrained model identifier from huggingface.co/models.",
)
parser.add_argument(
"--dataset_name",
type=str,
default=None,
help=(
"The name of the Dataset (from the HuggingFace hub) to train on (could be your own, possibly private,"
" dataset). It can also be a path pointing to a local copy of a dataset in your filesystem,"
" or to a folder containing files that 🤗 Datasets can understand."
),
)
parser.add_argument(
"--dataset_config_name",
type=str,
default=None,
help="The config of the Dataset, leave as None if there's only one config.",
)
parser.add_argument(
"--train_data_dir",
type=str,
default=None,
help=(
"A folder containing the training data. Folder contents must follow the structure described in"
" https://huggingface.co/docs/datasets/image_dataset#imagefolder. In particular, a `metadata.jsonl` file"
" must exist to provide the captions for the images. Ignored if `dataset_name` is specified."
),
)
parser.add_argument(
"--image_column", type=str, default="image", help="The column of the dataset containing an image."
)
parser.add_argument(
"--caption_column",
type=str,
default="text",
help="The column of the dataset containing a caption or a list of captions.",
)
parser.add_argument(
"--max_train_samples",
type=int,
default=None,
help=(
"For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
),
)
parser.add_argument(
"--output_dir",
type=str,
default="sd-model-finetuned",
help="The output directory where the model predictions and checkpoints will be written.",
)
parser.add_argument(
"--cache_dir",
type=str,
default=None,
help="The directory where the downloaded models and datasets will be stored.",
)
parser.add_argument("--seed", type=int, default=None, help="A seed for reproducible training.")
parser.add_argument(
"--resolution",
type=int,
default=512,
help=(
"The resolution for input images, all the images in the train/validation dataset will be resized to this"
" resolution"
),
)
parser.add_argument(
"--center_crop",
default=False,
action="store_true",
help=(
"Whether to center crop the input images to the resolution. If not set, the images will be randomly"
" cropped. The images will be resized to the resolution first before cropping."
),
)
parser.add_argument(
"--random_flip",
action="store_true",
help="whether to randomly flip images horizontally",
)
parser.add_argument(
"--train_batch_size", type=int, default=16, help="Batch size (per device) for the training dataloader."
)
parser.add_argument("--num_train_epochs", type=int, default=100)
parser.add_argument(
"--max_train_steps",
type=int,
default=None,
help="Total number of training steps to perform. If provided, overrides num_train_epochs.",
)
parser.add_argument(
"--gradient_accumulation_steps",
type=int,
default=1,
help="Number of updates steps to accumulate before performing a backward/update pass.",
)
parser.add_argument(
"--gradient_checkpointing",
action="store_true",
help="Whether or not to use gradient checkpointing to save memory at the expense of slower backward pass.",
)
parser.add_argument(
"--learning_rate",
type=float,
default=1e-4,
help="Initial learning rate (after the potential warmup period) to use.",
)
parser.add_argument(
"--scale_lr",
action="store_true",
default=False,
help="Scale the learning rate by the number of GPUs, gradient accumulation steps, and batch size.",
)
parser.add_argument(
"--lr_scheduler",
type=str,
default="constant",
help=(
'The scheduler type to use. Choose between ["linear", "cosine", "cosine_with_restarts", "polynomial",'
' "constant", "constant_with_warmup"]'
),
)
parser.add_argument(
"--lr_warmup_steps", type=int, default=500, help="Number of steps for the warmup in the lr scheduler."
)
parser.add_argument(
"--use_8bit_adam", action="store_true", help="Whether or not to use 8-bit Adam from bitsandbytes."
)
parser.add_argument(
"--allow_tf32",
action="store_true",
help=(
"Whether or not to allow TF32 on Ampere GPUs. Can be used to speed up training. For more information, see"
" https://pytorch.org/docs/stable/notes/cuda.html#tensorfloat-32-tf32-on-ampere-devices"
),
)
parser.add_argument("--use_ema", action="store_true", help="Whether to use EMA model.")
parser.add_argument(
"--non_ema_revision",
type=str,
default=None,
required=False,
help=(
"Revision of pretrained non-ema model identifier. Must be a branch, tag or git identifier of the local or"
" remote repository specified with --pretrained_model_name_or_path."
),
)
parser.add_argument(
"--dataloader_num_workers",
type=int,
default=0,
help=(
"Number of subprocesses to use for data loading. 0 means that the data will be loaded in the main process."
),
)
parser.add_argument("--adam_beta1", type=float, default=0.9, help="The beta1 parameter for the Adam optimizer.")
parser.add_argument("--adam_beta2", type=float, default=0.999, help="The beta2 parameter for the Adam optimizer.")
parser.add_argument("--adam_weight_decay", type=float, default=1e-2, help="Weight decay to use.")
parser.add_argument("--adam_epsilon", type=float, default=1e-08, help="Epsilon value for the Adam optimizer")
parser.add_argument("--max_grad_norm", default=1.0, type=float, help="Max gradient norm.")
parser.add_argument("--push_to_hub", action="store_true", help="Whether or not to push the model to the Hub.")
parser.add_argument("--hub_token", type=str, default=None, help="The token to use to push to the Model Hub.")
parser.add_argument(
"--hub_model_id",
type=str,
default=None,
help="The name of the repository to keep in sync with the local `output_dir`.",
)
parser.add_argument(
"--logging_dir",
type=str,
default="logs",
help=(
"[TensorBoard](https://www.tensorflow.org/tensorboard) log directory. Will default to"
" *output_dir/runs/**CURRENT_DATETIME_HOSTNAME***."
),
)
parser.add_argument(
"--mixed_precision",
type=str,
default=None,
choices=["no", "fp16", "bf16"],
help=(
"Whether to use mixed precision. Choose between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >="
" 1.10.and an Nvidia Ampere GPU. Default to the value of accelerate config of the current system or the"
" flag passed with the `accelerate.launch` command. Use this argument to override the accelerate config."
),
)
parser.add_argument(
"--report_to",
type=str,
default="tensorboard",
help=(
'The integration to report the results and logs to. Supported platforms are `"tensorboard"`'
' (default), `"wandb"` and `"comet_ml"`. Use `"all"` to report to all integrations.'
),
)
parser.add_argument("--local_rank", type=int, default=-1, help="For distributed training: local_rank")
parser.add_argument(
"--checkpointing_steps",
type=int,
default=500,
help=(
"Save a checkpoint of the training state every X updates. These checkpoints are only suitable for resuming"
" training using `--resume_from_checkpoint`."
),
)
parser.add_argument(
"--checkpoints_total_limit",
type=int,
default=None,
help=(
"Max number of checkpoints to store. Passed as `total_limit` to the `Accelerator` `ProjectConfiguration`."
" See Accelerator::save_state https://huggingface.co/docs/accelerate/package_reference/accelerator#accelerate.Accelerator.save_state"
" for more docs"
),
)
parser.add_argument(
"--resume_from_checkpoint",
type=str,
default=None,
help=(
"Whether training should be resumed from a previous checkpoint. Use a path saved by"
' `--checkpointing_steps`, or `"latest"` to automatically select the last available checkpoint.'
),
)
parser.add_argument(
"--enable_xformers_memory_efficient_attention", action="store_true", help="Whether or not to use xformers."
)
parser.add_argument("--noise_offset", type=float, default=0, help="The scale of noise offset.")
args = parser.parse_args()
env_local_rank = int(os.environ.get("LOCAL_RANK", -1))
if env_local_rank != -1 and env_local_rank != args.local_rank:
args.local_rank = env_local_rank
# Sanity checks
if args.dataset_name is None and args.train_data_dir is None:
raise ValueError("Need either a dataset name or a training folder.")
# default to using the same revision for the non-ema model if not specified
if args.non_ema_revision is None:
args.non_ema_revision = args.revision
return args
dataset_name_mapping = {
"lambdalabs/pokemon-blip-captions": ("image", "text"),
}
def main():
args = parse_args()
if args.non_ema_revision is not None:
deprecate(
"non_ema_revision!=None",
"0.15.0",
message=(
"Downloading 'non_ema' weights from revision branches of the Hub is deprecated. Please make sure to"
" use `--variant=non_ema` instead."
),
)
logging_dir = os.path.join(args.output_dir, args.logging_dir)
accelerator_project_config = ProjectConfiguration(total_limit=args.checkpoints_total_limit)
accelerator = Accelerator(
gradient_accumulation_steps=args.gradient_accumulation_steps,
mixed_precision=args.mixed_precision,
log_with=args.report_to,
logging_dir=logging_dir,
project_config=accelerator_project_config,
)
# Make one log on every process with the configuration for debugging.
logging.basicConfig(
format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
datefmt="%m/%d/%Y %H:%M:%S",
level=logging.INFO,
)
logger.info(accelerator.state, main_process_only=False)
if accelerator.is_local_main_process:
datasets.utils.logging.set_verbosity_warning()
transformers.utils.logging.set_verbosity_warning()
diffusers.utils.logging.set_verbosity_info()
else:
datasets.utils.logging.set_verbosity_error()
transformers.utils.logging.set_verbosity_error()
diffusers.utils.logging.set_verbosity_error()
# If passed along, set the training seed now.
if args.seed is not None:
set_seed(args.seed)
# Handle the repository creation
if accelerator.is_main_process:
if args.output_dir is not None:
os.makedirs(args.output_dir, exist_ok=True)
if args.push_to_hub:
repo_id = create_repo(
repo_id=args.hub_model_id or Path(args.output_dir).name, exist_ok=True, token=args.hub_token
).repo_id
# Load scheduler, tokenizer and models.
noise_scheduler = DDPMScheduler.from_pretrained(args.pretrained_model_name_or_path, subfolder="scheduler")
tokenizer = CLIPTokenizer.from_pretrained(
args.pretrained_model_name_or_path, subfolder="tokenizer", revision=args.revision
)
text_encoder = CLIPTextModel.from_pretrained(
args.pretrained_model_name_or_path, subfolder="text_encoder", revision=args.revision
)
vae = AutoencoderKL.from_pretrained(args.pretrained_model_name_or_path, subfolder="vae", revision=args.revision)
unet = UNet2DConditionModel.from_pretrained(
args.pretrained_model_name_or_path, subfolder="unet", revision=args.non_ema_revision
)
# Freeze vae and text_encoder
vae.requires_grad_(False)
text_encoder.requires_grad_(False)
# Create EMA for the unet.
if args.use_ema:
ema_unet = UNet2DConditionModel.from_pretrained(
args.pretrained_model_name_or_path, subfolder="unet", revision=args.revision
)
ema_unet = EMAModel(ema_unet.parameters(), model_cls=UNet2DConditionModel, model_config=ema_unet.config)
if args.enable_xformers_memory_efficient_attention:
if is_xformers_available():
import xformers
xformers_version = version.parse(xformers.__version__)
if xformers_version == version.parse("0.0.16"):
logger.warn(
"xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
)
unet.enable_xformers_memory_efficient_attention()
else:
raise ValueError("xformers is not available. Make sure it is installed correctly")
# `accelerate` 0.16.0 will have better support for customized saving
if version.parse(accelerate.__version__) >= version.parse("0.16.0"):
# create custom saving & loading hooks so that `accelerator.save_state(...)` serializes in a nice format
def save_model_hook(models, weights, output_dir):
if args.use_ema:
ema_unet.save_pretrained(os.path.join(output_dir, "unet_ema"))
for i, model in enumerate(models):
model.save_pretrained(os.path.join(output_dir, "unet"))
# make sure to pop weight so that corresponding model is not saved again
weights.pop()
def load_model_hook(models, input_dir):
if args.use_ema:
load_model = EMAModel.from_pretrained(os.path.join(input_dir, "unet_ema"), UNet2DConditionModel)
ema_unet.load_state_dict(load_model.state_dict())
ema_unet.to(accelerator.device)
del load_model
for i in range(len(models)):
# pop models so that they are not loaded again
model = models.pop()
# load diffusers style into model
load_model = UNet2DConditionModel.from_pretrained(input_dir, subfolder="unet")
model.register_to_config(**load_model.config)
model.load_state_dict(load_model.state_dict())
del load_model
accelerator.register_save_state_pre_hook(save_model_hook)
accelerator.register_load_state_pre_hook(load_model_hook)
if args.gradient_checkpointing:
unet.enable_gradient_checkpointing()
# Enable TF32 for faster training on Ampere GPUs,
# cf https://pytorch.org/docs/stable/notes/cuda.html#tensorfloat-32-tf32-on-ampere-devices
if args.allow_tf32:
torch.backends.cuda.matmul.allow_tf32 = True
if args.scale_lr:
args.learning_rate = (
args.learning_rate * args.gradient_accumulation_steps * args.train_batch_size * accelerator.num_processes
)
# Initialize the optimizer
if args.use_8bit_adam:
try:
import bitsandbytes as bnb
except ImportError:
raise ImportError(
"Please install bitsandbytes to use 8-bit Adam. You can do so by running `pip install bitsandbytes`"
)
optimizer_cls = bnb.optim.AdamW8bit
else:
optimizer_cls = torch.optim.AdamW
optimizer = optimizer_cls(
unet.parameters(),
lr=args.learning_rate,
betas=(args.adam_beta1, args.adam_beta2),
weight_decay=args.adam_weight_decay,
eps=args.adam_epsilon,
)
# Get the datasets: you can either provide your own training and evaluation files (see below)
# or specify a Dataset from the hub (the dataset will be downloaded automatically from the datasets Hub).
# In distributed training, the load_dataset function guarantees that only one local process can concurrently
# download the dataset.
if args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
dataset = load_dataset(
args.dataset_name,
args.dataset_config_name,
cache_dir=args.cache_dir,
)
else:
data_files = {}
if args.train_data_dir is not None:
data_files["train"] = os.path.join(args.train_data_dir, "**")
dataset = load_dataset(
"imagefolder",
data_files=data_files,
cache_dir=args.cache_dir,
)
# See more about loading custom images at
# https://huggingface.co/docs/datasets/v2.4.0/en/image_load#imagefolder
# Preprocessing the datasets.
# We need to tokenize inputs and targets.
column_names = dataset["train"].column_names
# 6. Get the column names for input/target.
dataset_columns = dataset_name_mapping.get(args.dataset_name, None)
if args.image_column is None:
image_column = dataset_columns[0] if dataset_columns is not None else column_names[0]
else:
image_column = args.image_column
if image_column not in column_names:
raise ValueError(
f"--image_column' value '{args.image_column}' needs to be one of: {', '.join(column_names)}"
)
if args.caption_column is None:
caption_column = dataset_columns[1] if dataset_columns is not None else column_names[1]
else:
caption_column = args.caption_column
if caption_column not in column_names:
raise ValueError(
f"--caption_column' value '{args.caption_column}' needs to be one of: {', '.join(column_names)}"
)
# Preprocessing the datasets.
# We need to tokenize input captions and transform the images.
def tokenize_captions(examples, is_train=True):
captions = []
for caption in examples[caption_column]:
if isinstance(caption, str):
captions.append(caption)
elif isinstance(caption, (list, np.ndarray)):
# take a random caption if there are multiple
captions.append(random.choice(caption) if is_train else caption[0])
else:
raise ValueError(
f"Caption column `{caption_column}` should contain either strings or lists of strings."
)
inputs = tokenizer(
captions, max_length=tokenizer.model_max_length, padding="max_length", truncation=True, return_tensors="pt"
)
return inputs.input_ids
# Preprocessing the datasets.
train_transforms = transforms.Compose(
[
transforms.Resize(args.resolution, interpolation=transforms.InterpolationMode.BILINEAR),
transforms.CenterCrop(args.resolution) if args.center_crop else transforms.RandomCrop(args.resolution),
transforms.RandomHorizontalFlip() if args.random_flip else transforms.Lambda(lambda x: x),
transforms.ToTensor(),
transforms.Normalize([0.5], [0.5]),
]
)
def preprocess_train(examples):
images = [image.convert("RGB") for image in examples[image_column]]
examples["pixel_values"] = [train_transforms(image) for image in images]
examples["input_ids"] = tokenize_captions(examples)
return examples
with accelerator.main_process_first():
if args.max_train_samples is not None:
dataset["train"] = dataset["train"].shuffle(seed=args.seed).select(range(args.max_train_samples))
# Set the training transforms
train_dataset = dataset["train"].with_transform(preprocess_train)
def collate_fn(examples):
pixel_values = torch.stack([example["pixel_values"] for example in examples])
pixel_values = pixel_values.to(memory_format=torch.contiguous_format).float()
input_ids = torch.stack([example["input_ids"] for example in examples])
return {"pixel_values": pixel_values, "input_ids": input_ids}
# DataLoaders creation:
train_dataloader = torch.utils.data.DataLoader(
train_dataset,
shuffle=True,
collate_fn=collate_fn,
batch_size=args.train_batch_size,
num_workers=args.dataloader_num_workers,
)
# Scheduler and math around the number of training steps.
overrode_max_train_steps = False
num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
if args.max_train_steps is None:
args.max_train_steps = args.num_train_epochs * num_update_steps_per_epoch
overrode_max_train_steps = True
lr_scheduler = get_scheduler(
args.lr_scheduler,
optimizer=optimizer,
num_warmup_steps=args.lr_warmup_steps * args.gradient_accumulation_steps,
num_training_steps=args.max_train_steps * args.gradient_accumulation_steps,
)
# Prepare everything with our `accelerator`.
unet, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
unet, optimizer, train_dataloader, lr_scheduler
)
if args.use_ema:
ema_unet.to(accelerator.device)
# For mixed precision training we cast the text_encoder and vae weights to half-precision
# as these models are only used for inference, keeping weights in full precision is not required.
weight_dtype = torch.float32
if accelerator.mixed_precision == "fp16":
weight_dtype = torch.float16
elif accelerator.mixed_precision == "bf16":
weight_dtype = torch.bfloat16
# Move text_encode and vae to gpu and cast to weight_dtype
text_encoder.to(accelerator.device, dtype=weight_dtype)
vae.to(accelerator.device, dtype=weight_dtype)
# We need to recalculate our total training steps as the size of the training dataloader may have changed.
num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
if overrode_max_train_steps:
args.max_train_steps = args.num_train_epochs * num_update_steps_per_epoch
# Afterwards we recalculate our number of training epochs
args.num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
# We need to initialize the trackers we use, and also store our configuration.
# The trackers initializes automatically on the main process.
if accelerator.is_main_process:
accelerator.init_trackers("text2image-fine-tune", config=vars(args))
# Train!
total_batch_size = args.train_batch_size * accelerator.num_processes * args.gradient_accumulation_steps
logger.info("***** Running training *****")
logger.info(f" Num examples = {len(train_dataset)}")
logger.info(f" Num Epochs = {args.num_train_epochs}")
logger.info(f" Instantaneous batch size per device = {args.train_batch_size}")
logger.info(f" Total train batch size (w. parallel, distributed & accumulation) = {total_batch_size}")
logger.info(f" Gradient Accumulation steps = {args.gradient_accumulation_steps}")
logger.info(f" Total optimization steps = {args.max_train_steps}")
global_step = 0
first_epoch = 0
# Potentially load in the weights and states from a previous save
if args.resume_from_checkpoint:
if args.resume_from_checkpoint != "latest":
path = os.path.basename(args.resume_from_checkpoint)
else:
# Get the most recent checkpoint
dirs = os.listdir(args.output_dir)
dirs = [d for d in dirs if d.startswith("checkpoint")]
dirs = sorted(dirs, key=lambda x: int(x.split("-")[1]))
path = dirs[-1] if len(dirs) > 0 else None
if path is None:
accelerator.print(
f"Checkpoint '{args.resume_from_checkpoint}' does not exist. Starting a new training run."
)
args.resume_from_checkpoint = None
else:
accelerator.print(f"Resuming from checkpoint {path}")
accelerator.load_state(os.path.join(args.output_dir, path))
global_step = int(path.split("-")[1])
resume_global_step = global_step * args.gradient_accumulation_steps
first_epoch = global_step // num_update_steps_per_epoch
resume_step = resume_global_step % (num_update_steps_per_epoch * args.gradient_accumulation_steps)
# Only show the progress bar once on each machine.
progress_bar = tqdm(range(global_step, args.max_train_steps), disable=not accelerator.is_local_main_process)
progress_bar.set_description("Steps")
for epoch in range(first_epoch, args.num_train_epochs):
unet.train()
train_loss = 0.0
for step, batch in enumerate(train_dataloader):
# Skip steps until we reach the resumed step
if args.resume_from_checkpoint and epoch == first_epoch and step < resume_step:
if step % args.gradient_accumulation_steps == 0:
progress_bar.update(1)
continue
with accelerator.accumulate(unet):
# Convert images to latent space
latents = vae.encode(batch["pixel_values"].to(weight_dtype)).latent_dist.sample()
latents = latents * vae.config.scaling_factor
# Sample noise that we'll add to the latents
noise = torch.randn_like(latents)
if args.noise_offset:
# https://www.crosslabs.org//blog/diffusion-with-offset-noise
noise += args.noise_offset * torch.randn(
(latents.shape[0], latents.shape[1], 1, 1), device=latents.device
)
bsz = latents.shape[0]
# Sample a random timestep for each image
timesteps = torch.randint(0, noise_scheduler.num_train_timesteps, (bsz,), device=latents.device)
timesteps = timesteps.long()
# Add noise to the latents according to the noise magnitude at each timestep
# (this is the forward diffusion process)
noisy_latents = noise_scheduler.add_noise(latents, noise, timesteps)
# Get the text embedding for conditioning
encoder_hidden_states = text_encoder(batch["input_ids"])[0]
# Get the target for loss depending on the prediction type
if noise_scheduler.config.prediction_type == "epsilon":
target = noise
elif noise_scheduler.config.prediction_type == "v_prediction":
target = noise_scheduler.get_velocity(latents, noise, timesteps)
else:
raise ValueError(f"Unknown prediction type {noise_scheduler.config.prediction_type}")
# Predict the noise residual and compute loss
model_pred = unet(noisy_latents, timesteps, encoder_hidden_states).sample
loss = F.mse_loss(model_pred.float(), target.float(), reduction="mean")
# Gather the losses across all processes for logging (if we use distributed training).
avg_loss = accelerator.gather(loss.repeat(args.train_batch_size)).mean()
train_loss += avg_loss.item() / args.gradient_accumulation_steps
# Backpropagate
accelerator.backward(loss)
if accelerator.sync_gradients:
accelerator.clip_grad_norm_(unet.parameters(), args.max_grad_norm)
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
# Checks if the accelerator has performed an optimization step behind the scenes
if accelerator.sync_gradients:
if args.use_ema:
ema_unet.step(unet.parameters())
progress_bar.update(1)
global_step += 1
accelerator.log({"train_loss": train_loss}, step=global_step)
train_loss = 0.0
if global_step % args.checkpointing_steps == 0:
if accelerator.is_main_process:
save_path = os.path.join(args.output_dir, f"checkpoint-{global_step}")
accelerator.save_state(save_path)
logger.info(f"Saved state to {save_path}")
logs = {"step_loss": loss.detach().item(), "lr": lr_scheduler.get_last_lr()[0]}
progress_bar.set_postfix(**logs)
if global_step >= args.max_train_steps:
break
# Create the pipeline using the trained modules and save it.
accelerator.wait_for_everyone()
if accelerator.is_main_process:
unet = accelerator.unwrap_model(unet)
if args.use_ema:
ema_unet.copy_to(unet.parameters())
pipeline = StableDiffusionPipeline.from_pretrained(
args.pretrained_model_name_or_path,
text_encoder=text_encoder,
vae=vae,
unet=unet,
revision=args.revision,
)
pipeline.save_pretrained(args.output_dir)
if args.push_to_hub:
upload_folder(
repo_id=repo_id,
folder_path=args.output_dir,
commit_message="End of training",
ignore_patterns=["step_*", "epoch_*"],
)
accelerator.end_training()
if __name__ == "__main__":
main()
|