Uploading remaining model files

convert_diffusers_to_original_stable_diffusion.py

@@ -0,0 +1,333 @@
+# Script for converting a HF Diffusers saved pipeline to a Stable Diffusion checkpoint.
+# *Only* converts the UNet, VAE, and Text Encoder.
+# Does not convert optimizer state or any other thing.
+
+import argparse
+import os.path as osp
+import re
+
+import torch
+from safetensors.torch import load_file, save_file
+
+
+# =================#
+# UNet Conversion #
+# =================#
+
+unet_conversion_map = [
+    # (stable-diffusion, HF Diffusers)
+    ("time_embed.0.weight", "time_embedding.linear_1.weight"),
+    ("time_embed.0.bias", "time_embedding.linear_1.bias"),
+    ("time_embed.2.weight", "time_embedding.linear_2.weight"),
+    ("time_embed.2.bias", "time_embedding.linear_2.bias"),
+    ("input_blocks.0.0.weight", "conv_in.weight"),
+    ("input_blocks.0.0.bias", "conv_in.bias"),
+    ("out.0.weight", "conv_norm_out.weight"),
+    ("out.0.bias", "conv_norm_out.bias"),
+    ("out.2.weight", "conv_out.weight"),
+    ("out.2.bias", "conv_out.bias"),
+]
+
+unet_conversion_map_resnet = [
+    # (stable-diffusion, HF Diffusers)
+    ("in_layers.0", "norm1"),
+    ("in_layers.2", "conv1"),
+    ("out_layers.0", "norm2"),
+    ("out_layers.3", "conv2"),
+    ("emb_layers.1", "time_emb_proj"),
+    ("skip_connection", "conv_shortcut"),
+]
+
+unet_conversion_map_layer = []
+# hardcoded number of downblocks and resnets/attentions...
+# would need smarter logic for other networks.
+for i in range(4):
+    # loop over downblocks/upblocks
+
+    for j in range(2):
+        # loop over resnets/attentions for downblocks
+        hf_down_res_prefix = f"down_blocks.{i}.resnets.{j}."
+        sd_down_res_prefix = f"input_blocks.{3*i + j + 1}.0."
+        unet_conversion_map_layer.append((sd_down_res_prefix, hf_down_res_prefix))
+
+        if i < 3:
+            # no attention layers in down_blocks.3
+            hf_down_atn_prefix = f"down_blocks.{i}.attentions.{j}."
+            sd_down_atn_prefix = f"input_blocks.{3*i + j + 1}.1."
+            unet_conversion_map_layer.append((sd_down_atn_prefix, hf_down_atn_prefix))
+
+    for j in range(3):
+        # loop over resnets/attentions for upblocks
+        hf_up_res_prefix = f"up_blocks.{i}.resnets.{j}."
+        sd_up_res_prefix = f"output_blocks.{3*i + j}.0."
+        unet_conversion_map_layer.append((sd_up_res_prefix, hf_up_res_prefix))
+
+        if i > 0:
+            # no attention layers in up_blocks.0
+            hf_up_atn_prefix = f"up_blocks.{i}.attentions.{j}."
+            sd_up_atn_prefix = f"output_blocks.{3*i + j}.1."
+            unet_conversion_map_layer.append((sd_up_atn_prefix, hf_up_atn_prefix))
+
+    if i < 3:
+        # no downsample in down_blocks.3
+        hf_downsample_prefix = f"down_blocks.{i}.downsamplers.0.conv."
+        sd_downsample_prefix = f"input_blocks.{3*(i+1)}.0.op."
+        unet_conversion_map_layer.append((sd_downsample_prefix, hf_downsample_prefix))
+
+        # no upsample in up_blocks.3
+        hf_upsample_prefix = f"up_blocks.{i}.upsamplers.0."
+        sd_upsample_prefix = f"output_blocks.{3*i + 2}.{1 if i == 0 else 2}."
+        unet_conversion_map_layer.append((sd_upsample_prefix, hf_upsample_prefix))
+
+hf_mid_atn_prefix = "mid_block.attentions.0."
+sd_mid_atn_prefix = "middle_block.1."
+unet_conversion_map_layer.append((sd_mid_atn_prefix, hf_mid_atn_prefix))
+
+for j in range(2):
+    hf_mid_res_prefix = f"mid_block.resnets.{j}."
+    sd_mid_res_prefix = f"middle_block.{2*j}."
+    unet_conversion_map_layer.append((sd_mid_res_prefix, hf_mid_res_prefix))
+
+
+def convert_unet_state_dict(unet_state_dict):
+    # buyer beware: this is a *brittle* function,
+    # and correct output requires that all of these pieces interact in
+    # the exact order in which I have arranged them.
+    mapping = {k: k for k in unet_state_dict.keys()}
+    for sd_name, hf_name in unet_conversion_map:
+        mapping[hf_name] = sd_name
+    for k, v in mapping.items():
+        if "resnets" in k:
+            for sd_part, hf_part in unet_conversion_map_resnet:
+                v = v.replace(hf_part, sd_part)
+            mapping[k] = v
+    for k, v in mapping.items():
+        for sd_part, hf_part in unet_conversion_map_layer:
+            v = v.replace(hf_part, sd_part)
+        mapping[k] = v
+    new_state_dict = {v: unet_state_dict[k] for k, v in mapping.items()}
+    return new_state_dict
+
+
+# ================#
+# VAE Conversion #
+# ================#
+
+vae_conversion_map = [
+    # (stable-diffusion, HF Diffusers)
+    ("nin_shortcut", "conv_shortcut"),
+    ("norm_out", "conv_norm_out"),
+    ("mid.attn_1.", "mid_block.attentions.0."),
+]
+
+for i in range(4):
+    # down_blocks have two resnets
+    for j in range(2):
+        hf_down_prefix = f"encoder.down_blocks.{i}.resnets.{j}."
+        sd_down_prefix = f"encoder.down.{i}.block.{j}."
+        vae_conversion_map.append((sd_down_prefix, hf_down_prefix))
+
+    if i < 3:
+        hf_downsample_prefix = f"down_blocks.{i}.downsamplers.0."
+        sd_downsample_prefix = f"down.{i}.downsample."
+        vae_conversion_map.append((sd_downsample_prefix, hf_downsample_prefix))
+
+        hf_upsample_prefix = f"up_blocks.{i}.upsamplers.0."
+        sd_upsample_prefix = f"up.{3-i}.upsample."
+        vae_conversion_map.append((sd_upsample_prefix, hf_upsample_prefix))
+
+    # up_blocks have three resnets
+    # also, up blocks in hf are numbered in reverse from sd
+    for j in range(3):
+        hf_up_prefix = f"decoder.up_blocks.{i}.resnets.{j}."
+        sd_up_prefix = f"decoder.up.{3-i}.block.{j}."
+        vae_conversion_map.append((sd_up_prefix, hf_up_prefix))
+
+# this part accounts for mid blocks in both the encoder and the decoder
+for i in range(2):
+    hf_mid_res_prefix = f"mid_block.resnets.{i}."
+    sd_mid_res_prefix = f"mid.block_{i+1}."
+    vae_conversion_map.append((sd_mid_res_prefix, hf_mid_res_prefix))
+
+
+vae_conversion_map_attn = [
+    # (stable-diffusion, HF Diffusers)
+    ("norm.", "group_norm."),
+    ("q.", "query."),
+    ("k.", "key."),
+    ("v.", "value."),
+    ("proj_out.", "proj_attn."),
+]
+
+
+def reshape_weight_for_sd(w):
+    # convert HF linear weights to SD conv2d weights
+    return w.reshape(*w.shape, 1, 1)
+
+
+def convert_vae_state_dict(vae_state_dict):
+    mapping = {k: k for k in vae_state_dict.keys()}
+    for k, v in mapping.items():
+        for sd_part, hf_part in vae_conversion_map:
+            v = v.replace(hf_part, sd_part)
+        mapping[k] = v
+    for k, v in mapping.items():
+        if "attentions" in k:
+            for sd_part, hf_part in vae_conversion_map_attn:
+                v = v.replace(hf_part, sd_part)
+            mapping[k] = v
+    new_state_dict = {v: vae_state_dict[k] for k, v in mapping.items()}
+    weights_to_convert = ["q", "k", "v", "proj_out"]
+    for k, v in new_state_dict.items():
+        for weight_name in weights_to_convert:
+            if f"mid.attn_1.{weight_name}.weight" in k:
+                print(f"Reshaping {k} for SD format")
+                new_state_dict[k] = reshape_weight_for_sd(v)
+    return new_state_dict
+
+
+# =========================#
+# Text Encoder Conversion #
+# =========================#
+
+
+textenc_conversion_lst = [
+    # (stable-diffusion, HF Diffusers)
+    ("resblocks.", "text_model.encoder.layers."),
+    ("ln_1", "layer_norm1"),
+    ("ln_2", "layer_norm2"),
+    (".c_fc.", ".fc1."),
+    (".c_proj.", ".fc2."),
+    (".attn", ".self_attn"),
+    ("ln_final.", "transformer.text_model.final_layer_norm."),
+    ("token_embedding.weight", "transformer.text_model.embeddings.token_embedding.weight"),
+    ("positional_embedding", "transformer.text_model.embeddings.position_embedding.weight"),
+]
+protected = {re.escape(x[1]): x[0] for x in textenc_conversion_lst}
+textenc_pattern = re.compile("|".join(protected.keys()))
+
+# Ordering is from https://github.com/pytorch/pytorch/blob/master/test/cpp/api/modules.cpp
+code2idx = {"q": 0, "k": 1, "v": 2}
+
+
+def convert_text_enc_state_dict_v20(text_enc_dict):
+    new_state_dict = {}
+    capture_qkv_weight = {}
+    capture_qkv_bias = {}
+    for k, v in text_enc_dict.items():
+        if (
+            k.endswith(".self_attn.q_proj.weight")
+            or k.endswith(".self_attn.k_proj.weight")
+            or k.endswith(".self_attn.v_proj.weight")
+        ):
+            k_pre = k[: -len(".q_proj.weight")]
+            k_code = k[-len("q_proj.weight")]
+            if k_pre not in capture_qkv_weight:
+                capture_qkv_weight[k_pre] = [None, None, None]
+            capture_qkv_weight[k_pre][code2idx[k_code]] = v
+            continue
+
+        if (
+            k.endswith(".self_attn.q_proj.bias")
+            or k.endswith(".self_attn.k_proj.bias")
+            or k.endswith(".self_attn.v_proj.bias")
+        ):
+            k_pre = k[: -len(".q_proj.bias")]
+            k_code = k[-len("q_proj.bias")]
+            if k_pre not in capture_qkv_bias:
+                capture_qkv_bias[k_pre] = [None, None, None]
+            capture_qkv_bias[k_pre][code2idx[k_code]] = v
+            continue
+
+        relabelled_key = textenc_pattern.sub(lambda m: protected[re.escape(m.group(0))], k)
+        new_state_dict[relabelled_key] = v
+
+    for k_pre, tensors in capture_qkv_weight.items():
+        if None in tensors:
+            raise Exception("CORRUPTED MODEL: one of the q-k-v values for the text encoder was missing")
+        relabelled_key = textenc_pattern.sub(lambda m: protected[re.escape(m.group(0))], k_pre)
+        new_state_dict[relabelled_key + ".in_proj_weight"] = torch.cat(tensors)
+
+    for k_pre, tensors in capture_qkv_bias.items():
+        if None in tensors:
+            raise Exception("CORRUPTED MODEL: one of the q-k-v values for the text encoder was missing")
+        relabelled_key = textenc_pattern.sub(lambda m: protected[re.escape(m.group(0))], k_pre)
+        new_state_dict[relabelled_key + ".in_proj_bias"] = torch.cat(tensors)
+
+    return new_state_dict
+
+
+def convert_text_enc_state_dict(text_enc_dict):
+    return text_enc_dict
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument("--model_path", default=None, type=str, required=True, help="Path to the model to convert.")
+    parser.add_argument("--checkpoint_path", default=None, type=str, required=True, help="Path to the output model.")
+    parser.add_argument("--half", action="store_true", help="Save weights in half precision.")
+    parser.add_argument(
+        "--use_safetensors", action="store_true", help="Save weights use safetensors, default is ckpt."
+    )
+
+    args = parser.parse_args()
+
+    assert args.model_path is not None, "Must provide a model path!"
+
+    assert args.checkpoint_path is not None, "Must provide a checkpoint path!"
+
+    # Path for safetensors
+    unet_path = osp.join(args.model_path, "unet", "diffusion_pytorch_model.safetensors")
+    vae_path = osp.join(args.model_path, "vae", "diffusion_pytorch_model.safetensors")
+    text_enc_path = osp.join(args.model_path, "text_encoder", "model.safetensors")
+
+    # Load models from safetensors if it exists, if it doesn't pytorch
+    if osp.exists(unet_path):
+        unet_state_dict = load_file(unet_path, device="cpu")
+    else:
+        unet_path = osp.join(args.model_path, "unet", "diffusion_pytorch_model.bin")
+        unet_state_dict = torch.load(unet_path, map_location="cpu")
+
+    if osp.exists(vae_path):
+        vae_state_dict = load_file(vae_path, device="cpu")
+    else:
+        vae_path = osp.join(args.model_path, "vae", "diffusion_pytorch_model.bin")
+        vae_state_dict = torch.load(vae_path, map_location="cpu")
+
+    if osp.exists(text_enc_path):
+        text_enc_dict = load_file(text_enc_path, device="cpu")
+    else:
+        text_enc_path = osp.join(args.model_path, "text_encoder", "pytorch_model.bin")
+        text_enc_dict = torch.load(text_enc_path, map_location="cpu")
+
+    # Convert the UNet model
+    unet_state_dict = convert_unet_state_dict(unet_state_dict)
+    unet_state_dict = {"model.diffusion_model." + k: v for k, v in unet_state_dict.items()}
+
+    # Convert the VAE model
+    vae_state_dict = convert_vae_state_dict(vae_state_dict)
+    vae_state_dict = {"first_stage_model." + k: v for k, v in vae_state_dict.items()}
+
+    # Easiest way to identify v2.0 model seems to be that the text encoder (OpenCLIP) is deeper
+    is_v20_model = "text_model.encoder.layers.22.layer_norm2.bias" in text_enc_dict
+
+    if is_v20_model:
+        # Need to add the tag 'transformer' in advance so we can knock it out from the final layer-norm
+        text_enc_dict = {"transformer." + k: v for k, v in text_enc_dict.items()}
+        text_enc_dict = convert_text_enc_state_dict_v20(text_enc_dict)
+        text_enc_dict = {"cond_stage_model.model." + k: v for k, v in text_enc_dict.items()}
+    else:
+        text_enc_dict = convert_text_enc_state_dict(text_enc_dict)
+        text_enc_dict = {"cond_stage_model.transformer." + k: v for k, v in text_enc_dict.items()}
+
+    # Put together new checkpoint
+    state_dict = {**unet_state_dict, **vae_state_dict, **text_enc_dict}
+    if args.half:
+        state_dict = {k: v.half() for k, v in state_dict.items()}
+
+    if args.use_safetensors:
+        save_file(state_dict, args.checkpoint_path)
+    else:
+        state_dict = {"state_dict": state_dict}
+        torch.save(state_dict, args.checkpoint_path)

model/model-001.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9adc95eeecdcba071ffe15f815c537ed350f899780da842d66fc2eb4f9626dd1
+size 2580250395

model/text_encoder/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "stabilityai/stable-diffusion-2",
+  "architectures": [
+    "CLIPTextModel"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "dropout": 0.0,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_size": 1024,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 23,
+  "pad_token_id": 1,
+  "projection_dim": 512,
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.2",
+  "vocab_size": 49408
+}

model/text_encoder/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ee0a15adae7f5b5a8e337f3a6ed108ad70bcfe6c550973b407fc48818b52d9f
+size 1361596304

model/unet/config.json

@@ -0,0 +1,72 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.22.0.dev0",
+  "_name_or_path": "stabilityai/stable-diffusion-2",
+  "act_fn": "silu",
+  "addition_embed_type": null,
+  "addition_embed_type_num_heads": 64,
+  "addition_time_embed_dim": null,
+  "attention_head_dim": [
+    5,
+    10,
+    20,
+    20
+  ],
+  "attention_type": "default",
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 1024,
+  "cross_attention_norm": null,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "dropout": 0.0,
+  "dual_cross_attention": false,
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 4,
+  "layers_per_block": 2,
+  "mid_block_only_cross_attention": null,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_attention_heads": null,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 4,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_out_scale_factor": 1.0,
+  "resnet_skip_time_act": false,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 96,
+  "time_cond_proj_dim": null,
+  "time_embedding_act_fn": null,
+  "time_embedding_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "transformer_layers_per_block": 1,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": true
+}

model/unet/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1191bdf19018935bdd5ce80abbb9bcaf799a4fac296e88d6f855bdf45ba1c9ae
+size 3463726504

model/vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2aa1f43011b553a4cba7f37456465cdbd48aab7b54b9348b890e8058ea7683ec
+size 334643268

train_dreambooth.py

@@ -0,0 +1,869 @@
+import argparse
+import hashlib
+import itertools
+import random
+import json
+import logging
+import math
+import os
+from contextlib import nullcontext
+from pathlib import Path
+from typing import Optional
+
+import torch
+import torch.nn.functional as F
+import torch.utils.checkpoint
+from torch.utils.data import Dataset
+
+from accelerate import Accelerator
+from accelerate.logging import get_logger
+from accelerate.utils import set_seed
+from diffusers import AutoencoderKL, DDIMScheduler, DDPMScheduler, StableDiffusionPipeline, UNet2DConditionModel
+from diffusers.optimization import get_scheduler
+from diffusers.utils.import_utils import is_xformers_available
+from huggingface_hub import HfFolder, Repository, whoami
+from PIL import Image
+from torchvision import transforms
+from tqdm.auto import tqdm
+from transformers import CLIPTextModel, CLIPTokenizer
+
+
+torch.backends.cudnn.benchmark = True
+
+
+logger = get_logger(__name__)
+
+
+def parse_args(input_args=None):
+    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(
+        "--pretrained_vae_name_or_path",
+        type=str,
+        default=None,
+        help="Path to pretrained vae or vae 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(
+        "--instance_data_dir",
+        type=str,
+        default=None,
+        help="A folder containing the training data of instance images.",
+    )
+    parser.add_argument(
+        "--class_data_dir",
+        type=str,
+        default=None,
+        help="A folder containing the training data of class images.",
+    )
+    parser.add_argument(
+        "--instance_prompt",
+        type=str,
+        default=None,
+        help="The prompt with identifier specifying the instance",
+    )
+    parser.add_argument(
+        "--class_prompt",
+        type=str,
+        default=None,
+        help="The prompt to specify images in the same class as provided instance images.",
+    )
+    parser.add_argument(
+        "--save_sample_prompt",
+        type=str,
+        default=None,
+        help="The prompt used to generate sample outputs to save.",
+    )
+    parser.add_argument(
+        "--save_sample_negative_prompt",
+        type=str,
+        default=None,
+        help="The negative prompt used to generate sample outputs to save.",
+    )
+    parser.add_argument(
+        "--n_save_sample",
+        type=int,
+        default=4,
+        help="The number of samples to save.",
+    )
+    parser.add_argument(
+        "--save_guidance_scale",
+        type=float,
+        default=7.5,
+        help="CFG for save sample.",
+    )
+    parser.add_argument(
+        "--save_infer_steps",
+        type=int,
+        default=20,
+        help="The number of inference steps for save sample.",
+    )
+    parser.add_argument(
+        "--pad_tokens",
+        default=False,
+        action="store_true",
+        help="Flag to pad tokens to length 77.",
+    )
+    parser.add_argument(
+        "--with_prior_preservation",
+        default=False,
+        action="store_true",
+        help="Flag to add prior preservation loss.",
+    )
+    parser.add_argument("--prior_loss_weight", type=float, default=1.0, help="The weight of prior preservation loss.")
+    parser.add_argument(
+        "--num_class_images",
+        type=int,
+        default=100,
+        help=(
+            "Minimal class images for prior preservation loss. If not have enough images, additional images will be"
+            " sampled with class_prompt."
+        ),
+    )
+    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_text_encoder", action="store_true", help="Whether to train the text encoder")
+    parser.add_argument(
+        "--train_batch_size", type=int, default=4, help="Batch size (per device) for the training dataloader."
+    )
+    parser.add_argument(
+        "--sample_batch_size", type=int, default=4, help="Batch size (per device) for sampling images."
+    )
+    parser.add_argument("--num_train_epochs", type=int, default=1)
+    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=5e-6,
+        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("--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("--log_interval", type=int, default=10, help="Log every N steps.")
+    parser.add_argument("--save_interval", type=int, default=10_000, help="Save weights every N steps.")
+    parser.add_argument("--save_min_steps", type=int, default=0, help="Start saving weights after N steps.")
+    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("--not_cache_latents", action="store_true", help="Do not precompute and cache latents from VAE.")
+    parser.add_argument("--hflip", action="store_true", help="Apply horizontal flip data augmentation.")
+    parser.add_argument("--local_rank", type=int, default=-1, help="For distributed training: local_rank")
+    parser.add_argument(
+        "--concepts_list",
+        type=str,
+        default=None,
+        help="Path to json containing multiple concepts, will overwrite parameters like instance_prompt, class_prompt, etc.",
+    )
+    parser.add_argument(
+        "--read_prompts_from_txts",
+        action="store_true",
+        help="Use prompt per image. Put prompts in the same directory as images, e.g. for image.png create image.png.txt.",
+    )
+
+    if input_args is not None:
+        args = parser.parse_args(input_args)
+    else:
+        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
+
+    return args
+
+
+class DreamBoothDataset(Dataset):
+    """
+    A dataset to prepare the instance and class images with the prompts for fine-tuning the model.
+    It pre-processes the images and the tokenizes prompts.
+    """
+
+    def __init__(
+        self,
+        concepts_list,
+        tokenizer,
+        with_prior_preservation=True,
+        size=512,
+        center_crop=False,
+        num_class_images=None,
+        pad_tokens=False,
+        hflip=False,
+        read_prompts_from_txts=False,
+    ):
+        self.size = size
+        self.center_crop = center_crop
+        self.tokenizer = tokenizer
+        self.with_prior_preservation = with_prior_preservation
+        self.pad_tokens = pad_tokens
+        self.read_prompts_from_txts = read_prompts_from_txts
+
+        self.instance_images_path = []
+        self.class_images_path = []
+
+        for concept in concepts_list:
+            inst_img_path = [
+                (x, concept["instance_prompt"])
+                for x in Path(concept["instance_data_dir"]).iterdir()
+                if x.is_file() and not str(x).endswith(".txt")
+            ]
+            self.instance_images_path.extend(inst_img_path)
+
+            if with_prior_preservation:
+                class_img_path = [(x, concept["class_prompt"]) for x in Path(concept["class_data_dir"]).iterdir() if x.is_file()]
+                self.class_images_path.extend(class_img_path[:num_class_images])
+
+        random.shuffle(self.instance_images_path)
+        self.num_instance_images = len(self.instance_images_path)
+        self.num_class_images = len(self.class_images_path)
+        self._length = max(self.num_class_images, self.num_instance_images)
+
+        self.image_transforms = transforms.Compose(
+            [
+                transforms.RandomHorizontalFlip(0.5 * hflip),
+                transforms.Resize(size, interpolation=transforms.InterpolationMode.BILINEAR),
+                transforms.CenterCrop(size) if center_crop else transforms.RandomCrop(size),
+                transforms.ToTensor(),
+                transforms.Normalize([0.5], [0.5]),
+            ]
+        )
+
+    def __len__(self):
+        return self._length
+
+    def __getitem__(self, index):
+        example = {}
+        instance_path, instance_prompt = self.instance_images_path[index % self.num_instance_images]
+
+        if self.read_prompts_from_txts:
+            with open(str(instance_path) + ".txt") as f:
+                instance_prompt = f.read().strip()
+
+        instance_image = Image.open(instance_path)
+        if not instance_image.mode == "RGB":
+            instance_image = instance_image.convert("RGB")
+
+        example["instance_images"] = self.image_transforms(instance_image)
+        example["instance_prompt_ids"] = self.tokenizer(
+            instance_prompt,
+            padding="max_length" if self.pad_tokens else "do_not_pad",
+            truncation=True,
+            max_length=self.tokenizer.model_max_length,
+        ).input_ids
+
+        if self.with_prior_preservation:
+            class_path, class_prompt = self.class_images_path[index % self.num_class_images]
+            class_image = Image.open(class_path)
+            if not class_image.mode == "RGB":
+                class_image = class_image.convert("RGB")
+            example["class_images"] = self.image_transforms(class_image)
+            example["class_prompt_ids"] = self.tokenizer(
+                class_prompt,
+                padding="max_length" if self.pad_tokens else "do_not_pad",
+                truncation=True,
+                max_length=self.tokenizer.model_max_length,
+            ).input_ids
+
+        return example
+
+
+class PromptDataset(Dataset):
+    "A simple dataset to prepare the prompts to generate class images on multiple GPUs."
+
+    def __init__(self, prompt, num_samples):
+        self.prompt = prompt
+        self.num_samples = num_samples
+
+    def __len__(self):
+        return self.num_samples
+
+    def __getitem__(self, index):
+        example = {}
+        example["prompt"] = self.prompt
+        example["index"] = index
+        return example
+
+
+class LatentsDataset(Dataset):
+    def __init__(self, latents_cache, text_encoder_cache):
+        self.latents_cache = latents_cache
+        self.text_encoder_cache = text_encoder_cache
+
+    def __len__(self):
+        return len(self.latents_cache)
+
+    def __getitem__(self, index):
+        return self.latents_cache[index], self.text_encoder_cache[index]
+
+
+class AverageMeter:
+    def __init__(self, name=None):
+        self.name = name
+        self.reset()
+
+    def reset(self):
+        self.sum = self.count = self.avg = 0
+
+    def update(self, val, n=1):
+        self.sum += val * n
+        self.count += n
+        self.avg = self.sum / self.count
+
+
+def get_full_repo_name(model_id: str, organization: Optional[str] = None, token: Optional[str] = None):
+    if token is None:
+        token = HfFolder.get_token()
+    if organization is None:
+        username = whoami(token)["name"]
+        return f"{username}/{model_id}"
+    else:
+        return f"{organization}/{model_id}"
+
+
+def main(args):
+    logging_dir = Path(args.output_dir, "0", args.logging_dir)
+
+    accelerator = Accelerator(
+        gradient_accumulation_steps=args.gradient_accumulation_steps,
+        mixed_precision=args.mixed_precision,
+        log_with="tensorboard",
+        project_dir=logging_dir,
+    )
+
+    logging.basicConfig(
+        format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
+        datefmt="%m/%d/%Y %H:%M:%S",
+        level=logging.INFO,
+    )
+
+    # Currently, it's not possible to do gradient accumulation when training two models with accelerate.accumulate
+    # This will be enabled soon in accelerate. For now, we don't allow gradient accumulation when training two models.
+    # TODO (patil-suraj): Remove this check when gradient accumulation with two models is enabled in accelerate.
+    if args.train_text_encoder and args.gradient_accumulation_steps > 1 and accelerator.num_processes > 1:
+        raise ValueError(
+            "Gradient accumulation is not supported when training the text encoder in distributed training. "
+            "Please set gradient_accumulation_steps to 1. This feature will be supported in the future."
+        )
+
+    if args.seed is not None:
+        set_seed(args.seed)
+
+    if args.concepts_list is None:
+        args.concepts_list = [
+            {
+                "instance_prompt": args.instance_prompt,
+                "class_prompt": args.class_prompt,
+                "instance_data_dir": args.instance_data_dir,
+                "class_data_dir": args.class_data_dir
+            }
+        ]
+    else:
+        with open(args.concepts_list, "r") as f:
+            args.concepts_list = json.load(f)
+
+    if args.with_prior_preservation:
+        pipeline = None
+        for concept in args.concepts_list:
+            class_images_dir = Path(concept["class_data_dir"])
+            class_images_dir.mkdir(parents=True, exist_ok=True)
+            cur_class_images = len(list(class_images_dir.iterdir()))
+
+            if cur_class_images < args.num_class_images:
+                torch_dtype = torch.float16 if accelerator.device.type == "cuda" else torch.float32
+                if pipeline is None:
+                    pipeline = StableDiffusionPipeline.from_pretrained(
+                        args.pretrained_model_name_or_path,
+                        vae=AutoencoderKL.from_pretrained(
+                            args.pretrained_vae_name_or_path or args.pretrained_model_name_or_path,
+                            subfolder=None if args.pretrained_vae_name_or_path else "vae",
+                            revision=None if args.pretrained_vae_name_or_path else args.revision,
+                            torch_dtype=torch_dtype
+                        ),
+                        torch_dtype=torch_dtype,
+                        safety_checker=None,
+                        revision=args.revision
+                    )
+                    pipeline.scheduler = DDIMScheduler.from_config(pipeline.scheduler.config)
+                    if is_xformers_available():
+                        pipeline.enable_xformers_memory_efficient_attention()
+                    pipeline.set_progress_bar_config(disable=True)
+                    pipeline.to(accelerator.device)
+
+                num_new_images = args.num_class_images - cur_class_images
+                logger.info(f"Number of class images to sample: {num_new_images}.")
+
+                sample_dataset = PromptDataset(concept["class_prompt"], num_new_images)
+                sample_dataloader = torch.utils.data.DataLoader(sample_dataset, batch_size=args.sample_batch_size)
+
+                sample_dataloader = accelerator.prepare(sample_dataloader)
+
+                with torch.autocast("cuda"), torch.inference_mode():
+                    for example in tqdm(
+                        sample_dataloader, desc="Generating class images", disable=not accelerator.is_local_main_process
+                    ):
+                        images = pipeline(
+                            example["prompt"],
+                            num_inference_steps=args.save_infer_steps
+                            ).images
+
+                        for i, image in enumerate(images):
+                            hash_image = hashlib.sha1(image.tobytes()).hexdigest()
+                            image_filename = class_images_dir / f"{example['index'][i] + cur_class_images}-{hash_image}.jpg"
+                            image.save(image_filename)
+
+        del pipeline
+        if torch.cuda.is_available():
+            torch.cuda.empty_cache()
+
+    # Load the tokenizer
+    if args.tokenizer_name:
+        tokenizer = CLIPTokenizer.from_pretrained(
+            args.tokenizer_name,
+            revision=args.revision,
+        )
+    elif args.pretrained_model_name_or_path:
+        tokenizer = CLIPTokenizer.from_pretrained(
+            args.pretrained_model_name_or_path,
+            subfolder="tokenizer",
+            revision=args.revision,
+        )
+
+    # Load models and create wrapper for stable diffusion
+    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,
+        torch_dtype=torch.float32
+    )
+
+    vae.requires_grad_(False)
+    if not args.train_text_encoder:
+        text_encoder.requires_grad_(False)
+
+    if is_xformers_available():
+        vae.enable_xformers_memory_efficient_attention()
+        unet.enable_xformers_memory_efficient_attention()
+    else:
+        logger.warning("xformers is not available. Make sure it is installed correctly")
+
+    if args.gradient_checkpointing:
+        unet.enable_gradient_checkpointing()
+        if args.train_text_encoder:
+            text_encoder.gradient_checkpointing_enable()
+
+    if args.scale_lr:
+        args.learning_rate = (
+            args.learning_rate * args.gradient_accumulation_steps * args.train_batch_size * accelerator.num_processes
+        )
+
+    # Use 8-bit Adam for lower memory usage or to fine-tune the model in 16GB GPUs
+    if args.use_8bit_adam:
+        try:
+            import bitsandbytes as bnb
+        except ImportError:
+            raise ImportError(
+                "To use 8-bit Adam, please install the bitsandbytes library: `pip install bitsandbytes`."
+            )
+
+        optimizer_class = bnb.optim.AdamW8bit
+    else:
+        optimizer_class = torch.optim.AdamW
+
+    params_to_optimize = (
+        itertools.chain(unet.parameters(), text_encoder.parameters()) if args.train_text_encoder else unet.parameters()
+    )
+    optimizer = optimizer_class(
+        params_to_optimize,
+        lr=args.learning_rate,
+        betas=(args.adam_beta1, args.adam_beta2),
+        weight_decay=args.adam_weight_decay,
+        eps=args.adam_epsilon,
+    )
+
+    noise_scheduler = DDPMScheduler.from_config(args.pretrained_model_name_or_path, subfolder="scheduler")
+
+    train_dataset = DreamBoothDataset(
+        concepts_list=args.concepts_list,
+        tokenizer=tokenizer,
+        with_prior_preservation=args.with_prior_preservation,
+        size=args.resolution,
+        center_crop=args.center_crop,
+        num_class_images=args.num_class_images,
+        pad_tokens=args.pad_tokens,
+        hflip=args.hflip,
+        read_prompts_from_txts=args.read_prompts_from_txts,
+    )
+
+    def collate_fn(examples):
+        input_ids = [example["instance_prompt_ids"] for example in examples]
+        pixel_values = [example["instance_images"] for example in examples]
+
+        # Concat class and instance examples for prior preservation.
+        # We do this to avoid doing two forward passes.
+        if args.with_prior_preservation:
+            input_ids += [example["class_prompt_ids"] for example in examples]
+            pixel_values += [example["class_images"] for example in examples]
+
+        pixel_values = torch.stack(pixel_values)
+        pixel_values = pixel_values.to(memory_format=torch.contiguous_format).float()
+
+        input_ids = tokenizer.pad(
+            {"input_ids": input_ids},
+            padding=True,
+            return_tensors="pt",
+        ).input_ids
+
+        batch = {
+            "input_ids": input_ids,
+            "pixel_values": pixel_values,
+        }
+        return batch
+
+    train_dataloader = torch.utils.data.DataLoader(
+        train_dataset, batch_size=args.train_batch_size, shuffle=True, collate_fn=collate_fn, pin_memory=True
+    )
+
+    weight_dtype = torch.float32
+    if args.mixed_precision == "fp16":
+        weight_dtype = torch.float16
+    elif args.mixed_precision == "bf16":
+        weight_dtype = torch.bfloat16
+
+    # Move text_encode and vae to gpu.
+    # 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.
+    vae.to(accelerator.device, dtype=weight_dtype)
+    if not args.train_text_encoder:
+        text_encoder.to(accelerator.device, dtype=weight_dtype)
+
+    if not args.not_cache_latents:
+        latents_cache = []
+        text_encoder_cache = []
+        for batch in tqdm(train_dataloader, desc="Caching latents"):
+            with torch.no_grad():
+                batch["pixel_values"] = batch["pixel_values"].to(accelerator.device, non_blocking=True, dtype=weight_dtype)
+                batch["input_ids"] = batch["input_ids"].to(accelerator.device, non_blocking=True)
+                latents_cache.append(vae.encode(batch["pixel_values"]).latent_dist)
+                if args.train_text_encoder:
+                    text_encoder_cache.append(batch["input_ids"])
+                else:
+                    text_encoder_cache.append(text_encoder(batch["input_ids"])[0])
+        train_dataset = LatentsDataset(latents_cache, text_encoder_cache)
+        train_dataloader = torch.utils.data.DataLoader(train_dataset, batch_size=1, collate_fn=lambda x: x, shuffle=True)
+
+        del vae
+        if not args.train_text_encoder:
+            del text_encoder
+        if torch.cuda.is_available():
+            torch.cuda.empty_cache()
+
+    # 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,
+    )
+
+    if args.train_text_encoder:
+        unet, text_encoder, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
+            unet, text_encoder, optimizer, train_dataloader, lr_scheduler
+        )
+    else:
+        unet, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
+            unet, optimizer, train_dataloader, lr_scheduler
+        )
+
+    # 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("dreambooth")
+
+    # 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 batches each epoch = {len(train_dataloader)}")
+    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}")
+
+    def save_weights(step):
+        # Create the pipeline using using the trained modules and save it.
+        if accelerator.is_main_process:
+            if args.train_text_encoder:
+                text_enc_model = accelerator.unwrap_model(text_encoder, keep_fp32_wrapper=True)
+            else:
+                text_enc_model = CLIPTextModel.from_pretrained(args.pretrained_model_name_or_path, subfolder="text_encoder", revision=args.revision)
+            pipeline = StableDiffusionPipeline.from_pretrained(
+                args.pretrained_model_name_or_path,
+                unet=accelerator.unwrap_model(unet, keep_fp32_wrapper=True),
+                text_encoder=text_enc_model,
+                vae=AutoencoderKL.from_pretrained(
+                    args.pretrained_vae_name_or_path or args.pretrained_model_name_or_path,
+                    subfolder=None if args.pretrained_vae_name_or_path else "vae",
+                    revision=None if args.pretrained_vae_name_or_path else args.revision,
+                ),
+                safety_checker=None,
+                torch_dtype=torch.float16,
+                revision=args.revision,
+            )
+            pipeline.scheduler = DDIMScheduler.from_config(pipeline.scheduler.config)
+            if is_xformers_available():
+                pipeline.enable_xformers_memory_efficient_attention()
+            save_dir = os.path.join(args.output_dir, f"{step}")
+            pipeline.save_pretrained(save_dir)
+            with open(os.path.join(save_dir, "args.json"), "w") as f:
+                json.dump(args.__dict__, f, indent=2)
+
+            if args.save_sample_prompt is not None:
+                pipeline = pipeline.to(accelerator.device)
+                g_cuda = torch.Generator(device=accelerator.device).manual_seed(args.seed)
+                pipeline.set_progress_bar_config(disable=True)
+                sample_dir = os.path.join(save_dir, "samples")
+                os.makedirs(sample_dir, exist_ok=True)
+                with torch.autocast("cuda"), torch.inference_mode():
+                    for i in tqdm(range(args.n_save_sample), desc="Generating samples"):
+                        images = pipeline(
+                            args.save_sample_prompt,
+                            negative_prompt=args.save_sample_negative_prompt,
+                            guidance_scale=args.save_guidance_scale,
+                            num_inference_steps=args.save_infer_steps,
+                            generator=g_cuda
+                        ).images
+                        images[0].save(os.path.join(sample_dir, f"{i}.png"))
+                del pipeline
+                if torch.cuda.is_available():
+                    torch.cuda.empty_cache()
+            print(f"[*] Weights saved at {save_dir}")
+
+    # Only show the progress bar once on each machine.
+    progress_bar = tqdm(range(args.max_train_steps), disable=not accelerator.is_local_main_process)
+    progress_bar.set_description("Steps")
+    global_step = 0
+    loss_avg = AverageMeter()
+    text_enc_context = nullcontext() if args.train_text_encoder else torch.no_grad()
+    for epoch in range(args.num_train_epochs):
+        unet.train()
+        if args.train_text_encoder:
+            text_encoder.train()
+        for step, batch in enumerate(train_dataloader):
+            with accelerator.accumulate(unet):
+                # Convert images to latent space
+                with torch.no_grad():
+                    if not args.not_cache_latents:
+                        latent_dist = batch[0][0]
+                    else:
+                        latent_dist = vae.encode(batch["pixel_values"].to(dtype=weight_dtype)).latent_dist
+                    latents = latent_dist.sample() * 0.18215
+
+                # 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
+                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)
+
+                # Get the text embedding for conditioning
+                with text_enc_context:
+                    if not args.not_cache_latents:
+                        if args.train_text_encoder:
+                            encoder_hidden_states = text_encoder(batch[0][1])[0]
+                        else:
+                            encoder_hidden_states = batch[0][1]
+                    else:
+                        encoder_hidden_states = text_encoder(batch["input_ids"])[0]
+
+                # 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}")
+
+                if args.with_prior_preservation:
+                    # Chunk the noise and model_pred into two parts and compute the loss on each part separately.
+                    model_pred, model_pred_prior = torch.chunk(model_pred, 2, dim=0)
+                    target, target_prior = torch.chunk(target, 2, dim=0)
+
+                    # Compute instance loss
+                    loss = F.mse_loss(model_pred.float(), target.float(), reduction="mean")
+
+                    # Compute prior loss
+                    prior_loss = F.mse_loss(model_pred_prior.float(), target_prior.float(), reduction="mean")
+
+                    # Add the prior loss to the instance loss.
+                    loss = loss + args.prior_loss_weight * prior_loss
+                else:
+                    loss = F.mse_loss(model_pred.float(), target.float(), reduction="mean")
+
+                accelerator.backward(loss)
+                # if accelerator.sync_gradients:
+                #     params_to_clip = (
+                #         itertools.chain(unet.parameters(), text_encoder.parameters())
+                #         if args.train_text_encoder
+                #         else unet.parameters()
+                #     )
+                #     accelerator.clip_grad_norm_(params_to_clip, args.max_grad_norm)
+                optimizer.step()
+                lr_scheduler.step()
+                optimizer.zero_grad(set_to_none=True)
+                loss_avg.update(loss.detach_(), bsz)
+
+            if not global_step % args.log_interval:
+                logs = {"loss": loss_avg.avg.item(), "lr": lr_scheduler.get_last_lr()[0]}
+                progress_bar.set_postfix(**logs)
+                accelerator.log(logs, step=global_step)
+
+            if global_step > 0 and not global_step % args.save_interval and global_step >= args.save_min_steps:
+                save_weights(global_step)
+
+            progress_bar.update(1)
+            global_step += 1
+
+            if global_step >= args.max_train_steps:
+                break
+
+        accelerator.wait_for_everyone()
+
+    save_weights(global_step)
+
+    accelerator.end_training()
+
+
+if __name__ == "__main__":
+    args = parse_args()
+    main(args)