CITATION.cff and fix F401 warning

CITATION.cff

@@ -0,0 +1,29 @@
+cff-version: 1.2.0
+title: The Alignment Handbook
+message: >-
+  Robust recipes to align language models with human and AI
+  preferences.
+type: software
+authors:
+  - given-names: Lewis
+    family-names: Tunstall
+  - given-names: Edward
+    family-names: Beeching
+  - given-names: Nathan
+    family-names: Lambert
+  - given-names: Nazneen
+    family-names: Rajani
+  - given-names: Shengyi
+    family-names: Huang
+  - given-names: Kashif
+    family-names: Rasul
+  - given-names: Alvaro
+    family-names: Bartolome
+  - given-names: Alexander
+    name-particle: M.
+    family-names: Rush
+  - given-names: Thomas
+    family-names: Wolf
+repository-code: 'https://github.com/huggingface/alignment-handbook'
+license: Apache-2.0
+version: 0.3.0.dev0

README.md

@@ -115,15 +115,14 @@ You can now check out the `scripts` and `recipes` directories for instructions o
 
 ## Citation
 
-If you find the content of this repo useful in your work, please cite it as follows:
+If you find the content of this repo useful in your work, please cite it as follows via '\usepackage{biblatex}`:
 
 ```bibtex
-@misc{alignment_handbook2023,
-  author = {Lewis Tunstall and Edward Beeching and Nathan Lambert and Nazneen Rajani and Shengyi Huang and Kashif Rasul and Alexander M. Rush and Thomas Wolf},
-  title = {The Alignment Handbook},
-  year = {2023},
-  publisher = {GitHub},
-  journal = {GitHub repository},
-  howpublished = {\url{https://github.com/huggingface/alignment-handbook}}
+@software{Tunstall_The_Alignment_Handbook,
+  author = {Tunstall, Lewis and Beeching, Edward and Lambert, Nathan and Rajani, Nazneen and Huang, Shengyi and Rasul, Kashif and Bartolome, Alvaro and M. Rush, Alexander and Wolf, Thomas},
+  license = {Apache-2.0},
+  title = {{The Alignment Handbook}},
+  url = {https://github.com/huggingface/alignment-handbook},
+  version = {0.3.0.dev0}
 }
 ```

src/alignment/__init__.py

@@ -11,3 +11,21 @@
     get_tokenizer,
     is_adapter_model,
 )
+
+
+__all__ = [
+    "DataArguments",
+    "DPOConfig",
+    "H4ArgumentParser",
+    "ModelArguments",
+    "SFTConfig",
+    "apply_chat_template",
+    "get_datasets",
+    "decontaminate_humaneval",
+    "get_checkpoint",
+    "get_kbit_device_map",
+    "get_peft_config",
+    "get_quantization_config",
+    "get_tokenizer",
+    "is_adapter_model",
+]

src/alignment/release.py

@@ -20,14 +20,29 @@
 
 
 REPLACE_PATTERNS = {
-    "init": (re.compile(r'^__version__\s+=\s+"([^"]+)"\s*$', re.MULTILINE), '__version__ = "VERSION"\n'),
-    "setup": (re.compile(r'^(\s*)version\s*=\s*"[^"]+",', re.MULTILINE), r'\1version="VERSION",'),
+    "init": (
+        re.compile(r'^__version__\s+=\s+"([^"]+)"\s*$', re.MULTILINE),
+        '__version__ = "VERSION"\n',
+    ),
+    "setup": (
+        re.compile(r'^(\s*)version\s*=\s*"[^"]+",', re.MULTILINE),
+        r'\1version="VERSION",',
+    ),
+    "citation": (re.compile(r"^version:\s+[^ ]+", re.MULTILINE), "version: VERSION"),
+    "readme": (
+        re.compile(r"version\s+=\s+\{[^}]+\}", re.MULTILINE),
+        "version = {VERSION}",
+    ),
 }
+
+README_FILE = "README.md"
+
 REPLACE_FILES = {
     "init": "src/alignment/__init__.py",
     "setup": "setup.py",
+    "citation": "CITATION.cff",
+    "readme": README_FILE,
 }
-README_FILE = "README.md"
 
 
 def update_version_in_file(fname, version, pattern):
@@ -95,7 +110,11 @@ def post_release_work():
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser()
-    parser.add_argument("--post_release", action="store_true", help="Whether this is pre or post release.")
+    parser.add_argument(
+        "--post_release",
+        action="store_true",
+        help="Whether this is pre or post release.",
+    )
     parser.add_argument("--patch", action="store_true", help="Whether or not this is a patch release.")
     args = parser.parse_args()
     if not args.post_release: