mr-philo/dreambooth-web-demo

Application of stable diffusion: presenting dreambooth using streamlit web demo

DreamBooth Web Demo

Environment

conda create -n dbweb python=3.10
conda activate dbweb
pip3 install torch torchvision torchaudio

Some basic requirements

pip3 install opencv-python streamlit

For diffusers

pip install --upgrade diffusers
pip install -U -r requirements.txt

To run dreambooth example with a web demo, run

streamlit run Streamlit-Welcome.py

If you want to train a model of your own, follow the original README.md ahead. Note that the model and output images will be presented as follows:

Dreambooth - web demo
|-- assets
|-- class-images
|-- pages
|-- Users                 
    |-- person1           # each model folder for each person
        |-- log           
        |-- model         # All model file
        |-- output        # Default folder for output images
        |-- train         # Folder for training images
    |-- person2
    |-- person3
    |-- ...
README.md
requirements.txt
...

For example, we prepare Bill Gates' training images in the example folder. We didn't provide the model due to the memory issue (about 5GB per model).

• To train the model, please refer train_dreambooth.py
• Check train.bash if you want to use this finetuning script
• To generate images from the trained model, please refer infer.py

original README.md from huggingface/diffusers/dreambooth

DreamBooth training example

DreamBooth is a method to personalize text2image models like stable diffusion given just a few(3~5) images of a subject. The train_dreambooth.py script shows how to implement the training procedure and adapt it for stable diffusion.

Running locally with PyTorch

Installing the dependencies

Before running the scripts, make sure to install the library's training dependencies:

pip install -U -r requirements.txt

And initialize an 🤗Accelerate environment with:

accelerate config

Dog toy example

You need to accept the model license before downloading or using the weights. In this example we'll use model version v1-4, so you'll need to visit its card, read the license and tick the checkbox if you agree.

You have to be a registered user in 🤗 Hugging Face Hub, and you'll also need to use an access token for the code to work. For more information on access tokens, please refer to this section of the documentation.

Run the following command to authenticate your token

huggingface-cli login

If you have already cloned the repo, then you won't need to go through these steps.

Now let's get our dataset. Download images from here and save them in a directory. This will be our training data.

And launch the training using

export MODEL_NAME="CompVis/stable-diffusion-v1-4"
export INSTANCE_DIR="path-to-instance-images"
export OUTPUT_DIR="path-to-save-model"

accelerate launch train_dreambooth.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --instance_data_dir=$INSTANCE_DIR \
  --output_dir=$OUTPUT_DIR \
  --instance_prompt="a photo of sks dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --gradient_accumulation_steps=1 \
  --learning_rate=5e-6 \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --max_train_steps=400

Training with prior-preservation loss

Prior-preservation is used to avoid overfitting and language-drift. Refer to the paper to learn more about it. For prior-preservation we first generate images using the model with a class prompt and then use those during training along with our data. According to the paper, it's recommended to generate num_epochs * num_samples images for prior-preservation. 200-300 works well for most cases.

export MODEL_NAME="CompVis/stable-diffusion-v1-4"
export INSTANCE_DIR="path-to-instance-images"
export CLASS_DIR="path-to-class-images"
export OUTPUT_DIR="path-to-save-model"

accelerate launch train_dreambooth.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --instance_data_dir=$INSTANCE_DIR \
  --class_data_dir=$CLASS_DIR \
  --output_dir=$OUTPUT_DIR \
  --with_prior_preservation --prior_loss_weight=1.0 \
  --instance_prompt="a photo of sks dog" \
  --class_prompt="a photo of dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --gradient_accumulation_steps=1 \
  --learning_rate=5e-6 \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --num_class_images=200 \
  --max_train_steps=800

Training on a 16GB GPU:

With the help of gradient checkpointing and the 8-bit optimizer from bitsandbytes it's possible to run train dreambooth on a 16GB GPU.

Install bitsandbytes with pip install bitsandbytes

export MODEL_NAME="CompVis/stable-diffusion-v1-4"
export INSTANCE_DIR="path-to-instance-images"
export CLASS_DIR="path-to-class-images"
export OUTPUT_DIR="path-to-save-model"

accelerate launch train_dreambooth.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --instance_data_dir=$INSTANCE_DIR \
  --class_data_dir=$CLASS_DIR \
  --output_dir=$OUTPUT_DIR \
  --with_prior_preservation --prior_loss_weight=1.0 \
  --instance_prompt="a photo of sks dog" \
  --class_prompt="a photo of dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --gradient_accumulation_steps=2 --gradient_checkpointing \
  --use_8bit_adam \
  --learning_rate=5e-6 \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --num_class_images=200 \
  --max_train_steps=800

Training on a 8 GB GPU:

By using DeepSpeed it's possible to offload some tensors from VRAM to either CPU or NVME allowing to train with less VRAM.

DeepSpeed needs to be enabled with accelerate config. During configuration answer yes to "Do you want to use DeepSpeed?". With DeepSpeed stage 2, fp16 mixed precision and offloading both parameters and optimizer state to cpu it's possible to train on under 8 GB VRAM with a drawback of requiring significantly more RAM (about 25 GB). See documentation for more DeepSpeed configuration options.

Changing the default Adam optimizer to DeepSpeed's special version of Adam deepspeed.ops.adam.DeepSpeedCPUAdam gives a substantial speedup but enabling it requires CUDA toolchain with the same version as pytorch. 8-bit optimizer does not seem to be compatible with DeepSpeed at the moment.

export MODEL_NAME="CompVis/stable-diffusion-v1-4"
export INSTANCE_DIR="path-to-instance-images"
export CLASS_DIR="path-to-class-images"
export OUTPUT_DIR="path-to-save-model"

accelerate launch train_dreambooth.py \
  --pretrained_model_name_or_path=$MODEL_NAME \
  --instance_data_dir=$INSTANCE_DIR \
  --class_data_dir=$CLASS_DIR \
  --output_dir=$OUTPUT_DIR \
  --with_prior_preservation --prior_loss_weight=1.0 \
  --instance_prompt="a photo of sks dog" \
  --class_prompt="a photo of dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --sample_batch_size=1 \
  --gradient_accumulation_steps=1 --gradient_checkpointing \
  --learning_rate=5e-6 \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --num_class_images=200 \
  --max_train_steps=800 \
  --mixed_precision=fp16

Fine-tune text encoder with the UNet.

The script also allows to fine-tune the text_encoder along with the unet. It's been observed experimentally that fine-tuning text_encoder gives much better results especially on faces. Pass the --train_text_encoder argument to the script to enable training text_encoder.

Note: Training text encoder requires more memory, with this option the training won't fit on 16GB GPU. It needs at least 24GB VRAM.

export MODEL_NAME="CompVis/stable-diffusion-v1-4"
export INSTANCE_DIR="path-to-instance-images"
export CLASS_DIR="path-to-class-images"
export OUTPUT_DIR="path-to-save-model"

accelerate launch train_dreambooth.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --train_text_encoder \
  --instance_data_dir=$INSTANCE_DIR \
  --class_data_dir=$CLASS_DIR \
  --output_dir=$OUTPUT_DIR \
  --with_prior_preservation --prior_loss_weight=1.0 \
  --instance_prompt="a photo of sks dog" \
  --class_prompt="a photo of dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --use_8bit_adam
  --gradient_checkpointing \
  --learning_rate=2e-6 \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --num_class_images=200 \
  --max_train_steps=800

Inference

Once you have trained a model using above command, the inference can be done simply using the StableDiffusionPipeline. Make sure to include the identifier(e.g. sks in above example) in your prompt.

from diffusers import StableDiffusionPipeline
import torch

model_id = "path-to-your-trained-model"
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")

prompt = "A photo of sks dog in a bucket"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7.5).images[0]

image.save("dog-bucket.png")

Running with Flax/JAX

For faster training on TPUs and GPUs you can leverage the flax training example. Follow the instructions above to get the model and dataset before running the script.

_Note: The flax example don't yet support features like gradient checkpoint, gradient accumulation etc, so to use flax for faster training we will need >30GB cards.

Before running the scripts, make sure to install the library's training dependencies:

pip install -U -r requirements_flax.txt

Training without prior preservation loss

export MODEL_NAME="duongna/stable-diffusion-v1-4-flax"
export INSTANCE_DIR="path-to-instance-images"
export OUTPUT_DIR="path-to-save-model"

python train_dreambooth_flax.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --instance_data_dir=$INSTANCE_DIR \
  --output_dir=$OUTPUT_DIR \
  --instance_prompt="a photo of sks dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --learning_rate=5e-6 \
  --max_train_steps=400

Training with prior preservation loss

export MODEL_NAME="duongna/stable-diffusion-v1-4-flax"
export INSTANCE_DIR="path-to-instance-images"
export CLASS_DIR="path-to-class-images"
export OUTPUT_DIR="path-to-save-model"

python train_dreambooth_flax.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --instance_data_dir=$INSTANCE_DIR \
  --class_data_dir=$CLASS_DIR \
  --output_dir=$OUTPUT_DIR \
  --with_prior_preservation --prior_loss_weight=1.0 \
  --instance_prompt="a photo of sks dog" \
  --class_prompt="a photo of dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --learning_rate=5e-6 \
  --num_class_images=200 \
  --max_train_steps=800

Fine-tune text encoder with the UNet.

export MODEL_NAME="duongna/stable-diffusion-v1-4-flax"
export INSTANCE_DIR="path-to-instance-images"
export CLASS_DIR="path-to-class-images"
export OUTPUT_DIR="path-to-save-model"

python train_dreambooth_flax.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --train_text_encoder \
  --instance_data_dir=$INSTANCE_DIR \
  --class_data_dir=$CLASS_DIR \
  --output_dir=$OUTPUT_DIR \
  --with_prior_preservation --prior_loss_weight=1.0 \
  --instance_prompt="a photo of sks dog" \
  --class_prompt="a photo of dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --learning_rate=2e-6 \
  --num_class_images=200 \
  --max_train_steps=800