Skip to content

Latest commit

 

History

History
149 lines (104 loc) · 4.12 KB

SIGMA.md

File metadata and controls

149 lines (104 loc) · 4.12 KB

PixArt Sigma Quickstart

In this example, we'll be training a PixArt Sigma model using the SimpleTuner toolkit and will be using the full model type, as it being a smaller model will likely fit in VRAM.

Prerequisites

Make sure that you have python installed. You can check this by running:

python --version

Installation

Clone the SimpleTuner repository and set up the python venv:

git clone --branch=release https://github.com/bghira/SimpleTuner.git

cd SimpleTuner

python -m venv .venv

source .venv/bin/activate

pip install -U poetry pip

Depending on your system, you will run one of 3 commands:

# MacOS
poetry install --no-root -C install/apple

# Linux
poetry install --no-root

# Linux with ROCM
poetry install --no-root -C install/rocm

Setting up the environment

To run SimpleTuner, you will need to set up a configuration file, the dataset and model directories, and a dataloader configuration file.

Configuration file

Copy sdxl-env.sh.example to sdxl-env.sh:

cp sdxl-env.sh.example sdxl-env.sh

There, you will need to modify the following variables:

  • MODEL_TYPE - Set this to full.
  • USE_BITFIT - Set this to false.
  • PIXART_SIGMA - Set this to true.
  • MODEL_NAME - Set this to PixArt-alpha/PixArt-Sigma-XL-2-1024-MS.
  • BASE_DIR - Set this to the directory where you want to store your outputs and datasets. It's recommended to use a full path here.

There are a few more if using a Mac M-series machine:

  • MIXED_PRECISION should be set to no.
  • USE_XFORMERS should be set to false.

Dataset considerations

It's crucial to have a substantial dataset to train your model on. There are limitations on the dataset size, and you will need to ensure that your dataset is large enough to train your model effectively. Note that the bare minimum dataset size is TRAIN_BATCH_SIZE * GRADIENT_ACCUMULATION_STEPS. The dataset will not be discoverable by the trainer if it is too small.

Depending on the dataset you have, you will need to set up your dataset directory and dataloader configuration file differently. In this example, we will be using pseudo-camera-10k as the dataset.

In your BASE_DIR directory, create a multidatabackend.json:

[
  {
    "id": "pseudo-camera-10k-pixart",
    "type": "local",
    "crop": true,
    "crop_aspect": "square",
    "crop_style": "center",
    "resolution": 0.5,
    "minimum_image_size": 0.25,
    "maximum_image_size": 1.0,
    "target_downsample_size": 1.0,
    "resolution_type": "area",
    "cache_dir_vae": "cache/vae/pixart/pseudo-camera-10k",
    "instance_data_dir": "datasets/pseudo-camera-10k",
    "disabled": false,
    "skip_file_discovery": "",
    "caption_strategy": "filename",
    "metadata_backend": "json"
  },
  {
    "id": "text-embeds",
    "type": "local",
    "dataset_type": "text_embeds",
    "default": true,
    "cache_dir": "cache/text/pixart/pseudo-camera-10k",
    "disabled": false,
    "write_batch_size": 128
  }
]

Then, navigate to the BASE_DIR directory and create a datasets directory:

apt -y install git-lfs
mkdir -p datasets
pushd datasets
    git clone https://huggingface.co/datasets/ptx0/pseudo-camera-10k
popd

This will download about 10k photograph samples to your datasets/pseudo-camera-10k directory, which will be automatically created for you.

Login to WandB and Huggingface Hub

You'll want to login to WandB and HF Hub before beginning training, especially if you're using PUSH_TO_HUB=true and --report_to=wandb.

If you're going to be pushing items to a Git LFS repository manually, you should also run git config --global credential.helper store

Run the following commands:

wandb login

and

huggingface-cli login

Follow the instructions to log in to both services.

Executing the training run

From the SimpleTuner directory, one simply has to run:

bash train_sdxl.sh

This will begin the text embed and VAE output caching to disk.

For more information, see the dataloader and tutorial documents.