textual_inversion_decomposed.py

# implementation is based on diffuser's implementation for textual inversion

"""
Copyright 2023 Google LLC

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

    https://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
limitations under the License.
"""

import argparse
import logging
import math
import os
import random
from pathlib import Path

import numpy as np
import PIL
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

# TODO: remove and import from diffusers.utils when the new version of diffusers is released
from packaging import version
from PIL import Image
from torch.utils.data import Dataset
from torchvision import transforms
from tqdm.auto import tqdm
from transformers import CLIPTextModel, CLIPTokenizer
import matplotlib.pyplot as plt
import glob

import diffusers
from diffusers import (
    AutoencoderKL,
    DDPMScheduler,
    DiffusionPipeline,
    DPMSolverMultistepScheduler,
    StableDiffusionPipeline,
    UNet2DConditionModel,
)
from diffusers.optimization import get_scheduler
from diffusers.utils import check_min_version, is_wandb_available
from diffusers.utils.import_utils import is_xformers_available
import wandb
from torch.utils.data import WeightedRandomSampler
from torchvision import transforms
import shutil
    

if version.parse(version.parse(PIL.__version__).base_version) >= version.parse("9.1.0"):
    PIL_INTERPOLATION = {
        "linear": PIL.Image.Resampling.BILINEAR,
        "bilinear": PIL.Image.Resampling.BILINEAR,
        "bicubic": PIL.Image.Resampling.BICUBIC,
        "lanczos": PIL.Image.Resampling.LANCZOS,
        "nearest": PIL.Image.Resampling.NEAREST,
    }
else:
    PIL_INTERPOLATION = {
        "linear": PIL.Image.LINEAR,
        "bilinear": PIL.Image.BILINEAR,
        "bicubic": PIL.Image.BICUBIC,
        "lanczos": PIL.Image.LANCZOS,
        "nearest": PIL.Image.NEAREST,
    }
# ------------------------------------------------------------------------------


# Will error if the minimal version of diffusers is not installed. Remove at your own risks.
# check_min_version("0.16.0.dev0")

logger = get_logger(__name__)


def log_validation(text_encoder, tokenizer, unet, vae, args, accelerator, weight_dtype, epoch, global_step):
    logger.info(
        f"Running validation... \n Generating {args.num_validation_images} images with prompt:"
        f" {args.validation_prompt}."
    )
    # create pipeline (note: unet and vae are loaded again in float32)
    pipeline = DiffusionPipeline.from_pretrained(
        args.pretrained_model_name_or_path,
        text_encoder=accelerator.unwrap_model(text_encoder),
        tokenizer=tokenizer,
        unet=unet,
        vae=vae,
        revision=args.revision,
        torch_dtype=weight_dtype,
        safety_checker=None, requires_safety_checker=False
    )
    pipeline.scheduler = DPMSolverMultistepScheduler.from_config(pipeline.scheduler.config)
    pipeline = pipeline.to(accelerator.device)
    pipeline.set_progress_bar_config(disable=True)

    # run inference
    generator = None if args.seed is None else torch.Generator(device=accelerator.device).manual_seed(args.seed)
    
    # width, height
    plt.figure(figsize=(12, 12))
    
    for l, val_prompt in enumerate(args.validation_prompt.split(",")):
        prompt = args.num_validation_images * [val_prompt]
        images = pipeline(prompt, num_inference_steps=25, generator=generator).images
        np_images = np.hstack([np.asarray(img) for img in images])
        plt.subplot(len(args.validation_prompt.split(",")),1, l + 1)
        plt.imshow(np_images)
        plt.axis("off")
        plt.title(val_prompt)
    plt.tight_layout()

    if args.report_to == "wandb":
        wandb.log({"validation": wandb.Image(plt)}, step=global_step)

    os.makedirs(f"{args.output_dir}/samples/", exist_ok=True)
    plt.savefig(f"{args.output_dir}/samples/{global_step}.jpg", bbox_inches='tight')
    plt.close()

    del pipeline
    torch.cuda.empty_cache()


def save_progress(text_encoder, placeholder_token_ids, accelerator, args, save_path):
    logger.info("Saving embeddings")
    learned_embeds_dict = {}
    for i in range(len(args.placeholder_token.split(" "))):
        learned_embeds = accelerator.unwrap_model(text_encoder).get_input_embeddings().weight[placeholder_token_ids[i]]
        learned_embeds_dict[args.placeholder_token.split(" ")[i]] = learned_embeds.detach().cpu()
    torch.save(learned_embeds_dict, save_path)


def parse_args():
    parser = argparse.ArgumentParser(description="Simple example of a training script.")
    parser.add_argument(
        "--save_steps",
        type=int,
        default=50,
        help="Save learned_embeds.bin every X updates steps.",
    )
    
    parser.add_argument(
        "--pretrained_model_name_or_path",
        type=str,
        default="runwayml/stable-diffusion-v1-5",
        required=False,
        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(
        "--tokenizer_name",
        type=str,
        default=None,
        help="Pretrained tokenizer name or path if not the same as model_name",
    )
    parser.add_argument(
        "--train_data_dir", type=str, default=None, required=True, help="A folder containing the training data."
    )
    parser.add_argument(
        "--placeholder_token",
        type=str,
        default=None,
        required=True,
        help="A token to use as a placeholder for the concept.",
    )
    parser.add_argument(
        "--initializer_token", type=str, default="object object", help="A token to use as initializer word."
    )
    parser.add_argument("--learnable_property", type=str, default="object", help="Choose between 'object' and 'style'")
    parser.add_argument("--repeats", type=int, default=100, help="How many times to repeat the training data.")
    parser.add_argument(
        "--output_dir",
        type=str,
        default="text-inversion-model",
        help="The output directory where the model predictions and checkpoints will be written.",
    )
    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", action="store_true", help="Whether to center crop images before resizing to resolution."
    )
    parser.add_argument(
        "--train_batch_size", type=int, default=2, 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=5000,
        help="Total number of training steps to perform.  If provided, overrides num_train_epochs.",
    )
    parser.add_argument(
        "--gradient_accumulation_steps",
        type=int,
        default=4,
        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=5.0e-04,
        help="Initial learning rate (after the potential warmup period) to use.",
    )
    parser.add_argument(
        "--scale_lr",
        action="store_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=0, help="Number of steps for the warmup in the lr scheduler."
    )
    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(
        "--mixed_precision",
        type=str,
        default="no",
        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."
        ),
    )
    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(
        "--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("--wandb_run_name", type=str, default="test")
    parser.add_argument("--wandb_user", type=str, default="none")
    parser.add_argument("--wandb_project_name", type=str, default="none")
     
    
    parser.add_argument(
        "--validation_prompt",
        type=str,
        default=None,
        help="A prompt that is used during validation to verify that the model is learning.",
    )
    parser.add_argument(
        "--num_validation_images",
        type=int,
        default=4,
        help="Number of images that should be generated during validation with `validation_prompt`.",
    )
    parser.add_argument(
        "--validation_steps",
        type=int,
        default=50,
        help=(
            "Run validation every X steps. Validation consists of running the prompt"
            " `args.validation_prompt` multiple times: `args.num_validation_images`"
            " and logging the images."
        ),
    )

    parser.add_argument("--local_rank", type=int, default=-1, help="For distributed training: local_rank")
    
    parser.add_argument(
        "--checkpointing_steps",
        type=int,
        default=200,
        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(
        "--t_dist", type=float, default=0.5
    )

    parser.add_argument(
        "--random_drop", type=float, default=0
    )

    parser.add_argument("--path_to_learned_embeds", type=str, default=None)
    parser.add_argument("--added_placeholders", type=str, default=None)
    parser.add_argument("--opt_placeholders", type=str, default=None)
    parser.add_argument("--freezed_placeholders", type=str, default=None)

    # clip refinement related
    parser.add_argument("--path_to_clip_selected", type=str, default=None)
    

    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

    if args.train_data_dir is None:
        raise ValueError("You must specify a train data directory.")
    
    # if args.initializer_token is None:
    #     # in this case, we do clip refinment
    #     assert os.path.exists(f"{args.path_to_clip_selected}")
    #     seed_selection_embds = torch.load(f"{args.path_to_clip_selected}")
    #     args.initializer_token = seed_selection_embds["initializer_token"]
    #     print(f"initializer_token {args.initializer_token} loaded ...") 
    return args


imagenet_templates_small = [
    "a photo of a {}",
    "a rendering of a {}",
    "a cropped photo of the {}",
    "the photo of a {}",
    "a photo of a clean {}",
    "a photo of a dirty {}",
    "a dark photo of the {}",
    "a photo of my {}",
    "a photo of the cool {}",
    "a close-up photo of a {}",
    "a bright photo of the {}",
    "a cropped photo of a {}",
    "a photo of the {}",
    "a good photo of the {}",
    "a photo of one {}",
    "a close-up photo of the {}",
    "a rendition of the {}",
    "a photo of the clean {}",
    "a rendition of a {}",
    "a photo of a nice {}",
    "a good photo of a {}",
    "a photo of the nice {}",
    "a photo of the small {}",
    "a photo of the weird {}",
    "a photo of the large {}",
    "a photo of a cool {}",
    "a photo of a small {}",
]

imagenet_style_templates_small = [
    "a painting in the style of {}",
    "a rendering in the style of {}",
    "a cropped painting in the style of {}",
    "the painting in the style of {}",
    "a clean painting in the style of {}",
    "a dirty painting in the style of {}",
    "a dark painting in the style of {}",
    "a picture in the style of {}",
    "a cool painting in the style of {}",
    "a close-up painting in the style of {}",
    "a bright painting in the style of {}",
    "a cropped painting in the style of {}",
    "a good painting in the style of {}",
    "a close-up painting in the style of {}",
    "a rendition in the style of {}",
    "a nice painting in the style of {}",
    "a small painting in the style of {}",
    "a weird painting in the style of {}",
    "a large painting in the style of {}",
]


class TextualInversionDataset(Dataset):
    def __init__(
        self,
        data_root,
        tokenizer,
        learnable_property="object",  # [object, style]
        size=512,
        repeats=100,
        interpolation="bicubic",
        flip_p=0.5,
        set="train",
        placeholder_token="*",
        center_crop=False,
    ):
        self.data_root = data_root
        self.tokenizer = tokenizer
        self.learnable_property = learnable_property
        self.size = size
        self.placeholder_token = placeholder_token
        self.center_crop = center_crop
        self.flip_p = flip_p

        self.image_paths = glob.glob(f"{self.data_root}/*.png") + glob.glob(f"{self.data_root}/*.jpg") + glob.glob(f"{self.data_root}/*.jpeg")

        self.num_images = len(self.image_paths)
        self._length = self.num_images

        if set == "train":
            self._length = self.num_images * repeats

        self.interpolation = {
            "linear": PIL_INTERPOLATION["linear"],
            "bilinear": PIL_INTERPOLATION["bilinear"],
            "bicubic": PIL_INTERPOLATION["bicubic"],
            "lanczos": PIL_INTERPOLATION["lanczos"],
        }[interpolation]

        self.templates = imagenet_style_templates_small if learnable_property == "style" else imagenet_templates_small
        self.flip_transform = transforms.RandomHorizontalFlip(p=self.flip_p)

    def __len__(self):
        return self._length

    def __getitem__(self, i):
        example = {}
        image = Image.open(self.image_paths[i % self.num_images])

        if not image.mode == "RGB":
            image = image.convert("RGB")

        placeholder_string = self.placeholder_token
        template = random.choice(self.templates)
        text = template.format(placeholder_string)

        example["input_ids"] = self.tokenizer(
            text,
            padding="max_length",
            truncation=True,
            max_length=self.tokenizer.model_max_length,
            return_tensors="pt",
        ).input_ids[0]

        if len(placeholder_string.split(" ")) > 1:
            text_left = template.format(placeholder_string.split(" ")[0])
            example["input_ids_left"] = self.tokenizer(
                text_left,
                padding="max_length",
                truncation=True,
                max_length=self.tokenizer.model_max_length,
                return_tensors="pt",
            ).input_ids[0]
            text_right = template.format(placeholder_string.split(" ")[1])
            example["input_ids_right"] = self.tokenizer(
                text_right,
                padding="max_length",
                truncation=True,
                max_length=self.tokenizer.model_max_length,
                return_tensors="pt",
            ).input_ids[0]

        # default to score-sde preprocessing
        img = np.array(image).astype(np.uint8)

        if self.center_crop:
            crop = min(img.shape[0], img.shape[1])
            (
                h,
                w,
            ) = (
                img.shape[0],
                img.shape[1],
            )
            img = img[(h - crop) // 2 : (h + crop) // 2, (w - crop) // 2 : (w + crop) // 2]

        image = Image.fromarray(img)
        image = image.resize((self.size, self.size), resample=self.interpolation)

        image = self.flip_transform(image)
        image = np.array(image).astype(np.uint8)
        image = (image / 127.5 - 1.0).astype(np.float32)

        example["pixel_values"] = torch.from_numpy(image).permute(2, 0, 1)
        return example


def main():
    args = parse_args()

    accelerator = Accelerator(
        gradient_accumulation_steps=args.gradient_accumulation_steps,
        mixed_precision=args.mixed_precision,
    )

    # 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:
        transformers.utils.logging.set_verbosity_warning()
        diffusers.utils.logging.set_verbosity_info()
    else:
        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)

    # Load tokenizer
    tokenizer = CLIPTokenizer.from_pretrained(args.pretrained_model_name_or_path, subfolder="tokenizer")

    # Load scheduler and models
    noise_scheduler = DDPMScheduler.from_pretrained(args.pretrained_model_name_or_path, subfolder="scheduler")
    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.revision
    )

    # Add the placeholder token in tokenizer
    added_placeholders = args.placeholder_token.split(" ") if args.added_placeholders is None else args.added_placeholders.split(" ")
    print("added_placeholders", added_placeholders)
    num_added_tokens = tokenizer.add_tokens(added_placeholders)
    if num_added_tokens == 0:
        raise ValueError(
            f"The tokenizer already contains the token {args.placeholder_token}. Please pass a different"
            " `placeholder_token` that is not already in the tokenizer."
        )
    print("num_added_tokens", num_added_tokens)
    # exit(1)
    # Convert the initializer_token, placeholder_token to ids
    initializer_token_ids = tokenizer.encode(args.initializer_token, add_special_tokens=False)
    print("initializer_token_ids", initializer_token_ids)
    placeholder_token_ids = tokenizer.convert_tokens_to_ids(args.placeholder_token.split(" "))
    print("placeholder_token_ids", placeholder_token_ids)

    # Resize the token embeddings as we are adding new special tokens to the tokenizer
    text_encoder.resize_token_embeddings(len(tokenizer))

    # Initialise the newly added placeholder token with the embeddings of the initializer token
    token_embeds = text_encoder.get_input_embeddings().weight.data

    # Initilize the pretrained tokens in case we refine the tree
    if args.path_to_learned_embeds is not None:
        added_placeholder_token_ids = tokenizer.convert_tokens_to_ids(added_placeholders)
        assert os.path.exists(f"{args.path_to_learned_embeds}")
        learned_embeds_dict = torch.load(f"{args.path_to_learned_embeds}")
        for placeholder, placeholder_id in zip(added_placeholders, added_placeholder_token_ids):
            print(placeholder, placeholder_id)
            token_embeds[placeholder_id] = learned_embeds_dict[placeholder].to(accelerator.device).to(dtype=torch.float16)

    
    # Initialise the newly added placeholder token with the embeddings of the initializer token
    for initializer_token_id_, placeholder_token_id_ in zip(initializer_token_ids, placeholder_token_ids):
        print(f"initializer_token_id_ {initializer_token_id_}, placeholder_token_id_ {placeholder_token_id_}")
        token_embeds[placeholder_token_id_] = token_embeds[initializer_token_id_]

    # we only update the token ids of the placeholder token every iteration
    token_ids_opt = placeholder_token_ids if args.opt_placeholders is None else tokenizer.convert_tokens_to_ids(args.opt_placeholders.split(" "))
    print("token_ids_opt",token_ids_opt)
    index_no_updates = torch.Tensor(np.logical_not(np.array([x in token_ids_opt for x in torch.arange(len(tokenizer))]))).bool()
    print("index_no_updates")
    print(index_no_updates[-10:])

    # Freeze vae and unet
    vae.requires_grad_(False)
    unet.requires_grad_(False)
    # Freeze all parameters except for the token embeddings in text encoder
    text_encoder.text_model.encoder.requires_grad_(False)
    text_encoder.text_model.final_layer_norm.requires_grad_(False)
    text_encoder.text_model.embeddings.position_embedding.requires_grad_(False)

    if args.gradient_checkpointing:
        # Keep unet in train mode if we are using gradient checkpointing to save memory.
        # The dropout cannot be != 0 so it doesn't matter if we are in eval or train mode.
        unet.train()
        text_encoder.gradient_checkpointing_enable()
        unet.enable_gradient_checkpointing()

    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")

    # 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
    optimizer = torch.optim.AdamW(
        text_encoder.get_input_embeddings().parameters(),  # only optimize the embeddings
        lr=args.learning_rate,
        betas=(args.adam_beta1, args.adam_beta2),
        weight_decay=args.adam_weight_decay,
        eps=args.adam_epsilon,
    )

    # Dataset and DataLoaders creation:
    train_dataset = TextualInversionDataset(
        data_root=args.train_data_dir,
        tokenizer=tokenizer,
        size=args.resolution,
        placeholder_token=args.placeholder_token,
        repeats=args.repeats,
        learnable_property=args.learnable_property,
        center_crop=args.center_crop,
        set="train",
    )
    train_dataloader = torch.utils.data.DataLoader(
        train_dataset, batch_size=args.train_batch_size, shuffle=True, 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`.
    text_encoder, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
        text_encoder, optimizer, train_dataloader, lr_scheduler
    )

    # For mixed precision training we cast the unet 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 vae and unet to device and cast to weight_dtype
    unet.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)

    if args.report_to == "wandb":
        wandb.init(project=args.wandb_project_name, entity=args.wandb_user,
                   config=args, name=args.wandb_run_name, id=wandb.util.generate_id())
    
    log_validation(text_encoder, tokenizer, unet, vae, args, accelerator, torch.float32, 0, 0)
    
    # 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:
        accelerator.print(f"Resuming from checkpoint {args.resume_from_checkpoint}")
        accelerator.load_state(args.resume_from_checkpoint)
        global_step = int(os.path.basename(args.resume_from_checkpoint).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")

    # keep original embeddings as reference
    orig_embeds_params = accelerator.unwrap_model(text_encoder).get_input_embeddings().weight.data.clone()

    if args.t_dist:
        def func(x):
            return (1 / noise_scheduler.config.num_train_timesteps) * (1 - args.t_dist * np.cos(np.pi * x / noise_scheduler.config.num_train_timesteps))
        prob_t_weights = [func(t_) for t_ in np.arange(noise_scheduler.config.num_train_timesteps)]
    
    for epoch in range(first_epoch, args.num_train_epochs):
        text_encoder.train()
        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(text_encoder):
                # Convert images to latent space
                latents = vae.encode(batch["pixel_values"].to(dtype=weight_dtype)).latent_dist.sample().detach()
                latents = latents * vae.config.scaling_factor

                # Sample noise that we'll add to the latents
                noise = torch.randn_like(latents)
                bsz = latents.shape[0]
                # Sample a random timestep for each image
                if args.t_dist:
                    timesteps = torch.tensor(list(WeightedRandomSampler(prob_t_weights, bsz, replacement=True)), device=latents.device)
                else:
                    timesteps = torch.randint(0, noise_scheduler.config.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)

                ids_prompt_key = "input_ids"
                if args.random_drop:
                    if np.random.uniform(low=0, high=1) > args.random_drop:
                        ids_prompt_key = random.choice(["input_ids_left", "input_ids_right"])

                # Get the text embedding for conditioning
                encoder_hidden_states = text_encoder(batch[ids_prompt_key])[0].to(dtype=weight_dtype)
                # Predict the noise residual
                model_pred = unet(noisy_latents, timesteps, encoder_hidden_states).sample

                # 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}")

                loss = F.mse_loss(model_pred.float(), target.float(), reduction="mean")

                accelerator.backward(loss)

                optimizer.step()
                lr_scheduler.step()
                optimizer.zero_grad()

                # Let's make sure we don't update any embedding weights besides the newly added token
                with torch.no_grad():
                    accelerator.unwrap_model(text_encoder).get_input_embeddings().weight[
                        index_no_updates
                    ] = orig_embeds_params[index_no_updates]
                
                # Log norm
                with torch.no_grad():
                    norm_list = []
                    for placeholder_token_id_ in placeholder_token_ids:
                        cur_norm = torch.norm(accelerator.unwrap_model(text_encoder).get_input_embeddings().weight[placeholder_token_id_].unsqueeze(0))
                        norm_list.append(cur_norm)


            # Checks if the accelerator has performed an optimization step behind the scenes
            if accelerator.sync_gradients:
                progress_bar.update(1)
                global_step += 1
                if global_step % args.save_steps == 0:
                    save_path = os.path.join(args.output_dir, f"learned_embeds-steps-{global_step}.bin")
                    save_progress(text_encoder, placeholder_token_ids, accelerator, args, save_path)
                
                if accelerator.is_main_process:
                    if global_step % args.checkpointing_steps == 0:
                        # _before_ saving state, check if this save would set us over the `checkpoints_total_limit`
                        if args.checkpoints_total_limit is not None:
                            checkpoints = os.listdir(args.output_dir)
                            checkpoints = [d for d in checkpoints if d.startswith("checkpoint")]
                            checkpoints = sorted(checkpoints, key=lambda x: int(x.split("-")[1]))

                            # before we save the new checkpoint, we need to have at _most_ `checkpoints_total_limit - 1` checkpoints
                            if len(checkpoints) >= args.checkpoints_total_limit:
                                num_to_remove = len(checkpoints) - args.checkpoints_total_limit + 1
                                removing_checkpoints = checkpoints[0:num_to_remove]

                                logger.info(
                                    f"{len(checkpoints)} checkpoints already exist, removing {len(removing_checkpoints)} checkpoints"
                                )
                                logger.info(f"removing checkpoints: {', '.join(removing_checkpoints)}")

                                for removing_checkpoint in removing_checkpoints:
                                    removing_checkpoint = os.path.join(args.output_dir, removing_checkpoint)
                                    shutil.rmtree(removing_checkpoint)

                        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}")

                if args.validation_prompt is not None and global_step % args.validation_steps == 0:
                    log_validation(text_encoder, tokenizer, unet, vae, args, accelerator, weight_dtype, epoch, global_step)

            logs = {"loss": loss.detach().item(), "lr": lr_scheduler.get_last_lr()[0]}
            for k, token_ in enumerate(args.placeholder_token.split(" ")):
                logs[f"norm {token_}"] = norm_list[k].detach().item()
            progress_bar.set_postfix(**logs)
            if args.report_to == "wandb":
                wandb.log(logs, step=global_step)

            if global_step >= args.max_train_steps:
                break
    if args.report_to == "wandb":
        wandb.finish
    # Create the pipeline using the trained modules and save it.
    accelerator.wait_for_everyone()
    if accelerator.is_main_process:
        # Save the newly trained embeddings
        save_path = os.path.join(args.output_dir, "learned_embeds.bin")
        save_progress(text_encoder, placeholder_token_ids, accelerator, args, save_path)

    # accelerator.end_training()


if __name__ == "__main__":
    main()