Merge pull request #342 from NicholasCao/master

Add implementation of BeautifulPrompt and X-STA

examples/BeautifulPrompt/.gitignore

@@ -0,0 +1,152 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+docs/.build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# IDE
+.idea/
+.vscode/
+
+# macos
+*.DS_Store
+#data/
+
+docs/.build
+
+# pytorch checkpoint
+*.pt
+
+# wandb log
+example/wandb/
+
+outputs/*
+logs/*
+data/*
+!data/README.md
+wandb/*

examples/BeautifulPrompt/README.md

@@ -0,0 +1,43 @@
+# BeautifulPrompt
+This project is implemented for the EMNLP Industry Track 2023 paper: "BeautifulPrompt: Towards Automatic Prompt Engineering for Text-to-Image Synthesis". Our code is based on pytorch and huggingface transformers.
+
+## Data & Models
+We released our collected [dataset](https://atp-modelzoo-sh.oss-cn-shanghai.aliyuncs.com/release/tutorials/BeautifulPrompt/data.json), which includes prompt pairs and various scores, and also released a more extensive [rm_aesthetic dataset](https://atp-modelzoo-sh.oss-cn-shanghai.aliyuncs.com/release/tutorials/BeautifulPrompt/rm_aesthetic.json).
+
+We released the following models:
+- [alibaba-pai/pai-bloom-1b1-text2prompt-sd](https://huggingface.co/alibaba-pai/pai-bloom-1b1-text2prompt-sd)
+- [alibaba-pai/pai-bloom-1b1-text2prompt-sd-v2](https://huggingface.co/alibaba-pai/pai-bloom-1b1-text2prompt-sd-v2)
+
+## Run
+### Installation
+```bash
+conda create -n trlx python=3.10
+conda activate trlx
+
+pip install -e .
+pip install tensorboardX==2.6.0
+```
+
+### Training
+```bash
+# Step 1
+bash scripts/sft.sh
+# Step 2
+bash scripts/rm_aes.sh
+bash scripts/rm_ps.sh
+# Step 3
+bash scripts/ppo.sh
+```
+
+### Evaluation
+```bash
+pip install pillow
+pip install image-reward==1.2
+pip install git+https://github.com/openai/CLIP
+pip install diffusers
+
+bash scripts/eval.sh
+```
+
+## Acknowledgement
+This repo benefits from [trlx](https://github.com/CarperAI/trlx). Thanks for their wonderful works.