examples/research_projects/colossalai/README.md

# [DreamBooth](https://github.com/huggingface/diffusers/tree/main/examples/dreambooth) by [colossalai](https://github.com/hpcaitech/ColossalAI.git)

[DreamBooth](https://arxiv.org/abs/2208.12242) is a method to personalize text2image models like stable diffusion given just a few(3~5) images of a subject.
The `train_dreambooth_colossalai.py` script shows how to implement the training procedure and adapt it for stable diffusion.

By accommodating model data in CPU and GPU and moving the data to the computing device when necessary, [Gemini](https://www.colossalai.org/docs/advanced_tutorials/meet_gemini), the Heterogeneous Memory Manager of [Colossal-AI](https://github.com/hpcaitech/ColossalAI) can breakthrough the GPU memory wall by using GPU and CPU memory (composed of CPU DRAM or nvme SSD memory) together at the same time. Moreover, the model scale can be further improved by combining heterogeneous training with the other parallel approaches, such as data parallel, tensor parallel and pipeline parallel.

## Installing the dependencies

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

```bash
pip install -r requirements.txt
```

## Install [ColossalAI](https://github.com/hpcaitech/ColossalAI.git)

**From PyPI**
```bash
pip install colossalai
```

**From source**

```bash
git clone https://github.com/hpcaitech/ColossalAI.git
cd ColossalAI

# install colossalai
pip install .
```

## Dataset for Teyvat BLIP captions
Dataset used to train [Teyvat characters text to image model](https://github.com/hpcaitech/ColossalAI/tree/main/examples/images/diffusion).

BLIP generated captions for characters images from [genshin-impact fandom wiki](https://genshin-impact.fandom.com/wiki/Character#Playable_Characters)and [biligame wiki for genshin impact](https://wiki.biligame.com/ys/%E8%A7%92%E8%89%B2).

For each row the dataset contains `image` and `text` keys. `image` is a varying size PIL png, and `text` is the accompanying text caption. Only a train split is provided.

The `text` include the tag `Teyvat`, `Name`,`Element`, `Weapon`, `Region`, `Model type`, and `Description`, the `Description` is captioned with the [pre-trained BLIP model](https://github.com/salesforce/BLIP).

## Training

The argument `placement` can be `cpu`, `auto`, `cuda`, with `cpu` the GPU RAM required can be minimized to 4GB but will deceleration, with `cuda` you can also reduce GPU memory by half but accelerated training， with `auto` a more balanced solution for speed and memory can be obtained。

**___Note: Change the `resolution` to 768 if you are using the [stable-diffusion-2](https://huggingface.co/stabilityai/stable-diffusion-2) 768x768 model.___**

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

torchrun --nproc_per_node 2 train_dreambooth_colossalai.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 \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --max_train_steps=400 \
  --placement="cuda"
```


### 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. The `num_class_images` flag sets the number of images to generate with the class prompt. You can place existing images in `class_data_dir`, and the training script will generate any additional images so that `num_class_images` are present in `class_data_dir` during training time.

```bash
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"

torchrun --nproc_per_node 2 train_dreambooth_colossalai.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 \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --max_train_steps=800 \
  --placement="cuda"
```

## 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.

```python
from diffusers import StableDiffusionPipeline
import torch

model_id = "path-to-save-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")
```