run pre-commit

.github/workflows/python_ci.yml

@@ -5,7 +5,7 @@ on:
     branches: [main]
   push:
     branches: [main]
-  
+
 permissions:
   contents: read
 
@@ -19,7 +19,7 @@ jobs:
       with:
         python-version: "3.10"
     - uses: pre-commit/[email protected]
-  
+
   tests:
     name: Run the tests
     runs-on: ubuntu-latest
@@ -58,47 +58,47 @@ jobs:
     defaults:
       run:
         working-directory: bindings/python
-  
+
     steps:
       - uses: actions/checkout@v3
         with:
           fetch-depth: 0
-  
+
       - uses: actions/setup-python@v4
         with:
           cache: pip
           python-version: "3.11"
-  
+
       - name: Set up environment
         run: |
           pip install --upgrade "coverage[toml]>=5.1" diff-cover
-  
+
       - uses: actions/download-artifact@v3
         with:
           name: coverage-data
-  
+
       - name: Fetch master for coverage diff
         run: |
           git fetch --no-tags --prune origin main
-  
+
       - name: Combine coverage & fail if it's <100%.
         run: |
           # Combine coverage files (not needed now, but maybe later)
           # python -m coverage combine
-  
+
           # Produce an html report with absolute coverage information
           cd bindgins/python && python -m coverage html --skip-covered --skip-empty
-  
+
           # Report relative coverage and write to the workflow's summary
           cd bindings/python && python -m coverage xml
           diff-cover coverage.xml --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
-  
+
       - name: Upload HTML report if check failed.
         uses: actions/upload-artifact@v3
         with:
           name: html-report
           path: htmlcov
-        if: ${{ failure() }}    
+        if: ${{ failure() }}
 
   linux:
     runs-on: ${{ matrix.platform.runner }}
@@ -320,4 +320,4 @@ jobs:
 #           PYPI_TOKEN: ${{ secrets.PYPI_API_TOKEN }}
 #         with:
 #           command: upload
-#           args: --non-interactive --skip-existing wheels-*/*
+#           args: --non-interactive --skip-existing wheels-*/*

Cargo.toml

@@ -4,4 +4,4 @@
         "bindings/python",
     ]
 
-    resolver = "2"
+    resolver = "2"

bindings/python/Cargo.toml

@@ -10,4 +10,4 @@ crate-type = ["cdylib"]
 
 [dependencies]
 pyo3 = "0.22.0"
-outlines-core = { path = "../../outlines-core" }
+outlines-core = { path = "../../outlines-core" }

bindings/python/outlines_core/__init__.py

@@ -1,5 +1,5 @@
 """Outlines is a Generative Model Programming Framework."""
 import outlines_core.models
-from _outlines_core_rs import __version__ 
+from _outlines_core_rs import __version__
 
-__all__ = ["models", "__version__"]
+__all__ = ["models", "__version__"]

bindings/python/outlines_core/fsm/fsm.py

@@ -44,4 +44,4 @@ def allowed_token_ids(self, state: FSMState) -> Optional[Iterable[int]]:
         return next_instruction.tokens
 
     def next_state(self, state: FSMState, token_id: int) -> FSMState:
-        return FSMState(self.get_next_state(state, token_id))
+        return FSMState(self.get_next_state(state, token_id))

bindings/python/outlines_core/fsm/guide.py

@@ -292,4 +292,4 @@ def is_final_state(self, state: int) -> bool:
         return state in self.final_states
 
     def copy(self):
-        return self
+        return self

bindings/python/outlines_core/fsm/json_schema.py

@@ -516,4 +516,4 @@ def get_schema_from_signature(fn: Callable) -> str:
         )
     model = create_model(fn_name, **arguments)
 
-    return model.model_json_schema()
+    return model.model_json_schema()

bindings/python/outlines_core/fsm/regex.py

@@ -14,14 +14,6 @@
     cast,
 )
 
-from interegular.fsm import (
-    FSM,
-    Alphabet,
-    State,
-    TransitionKey,
-    _AnythingElseCls,
-    anything_else,
-)
 from _outlines_core_rs import (  # noqa: F401
     FSMInfo,
     _walk_fsm,
@@ -30,6 +22,14 @@
     get_vocabulary_transition_keys,
     state_scan_tokens,
 )
+from interegular.fsm import (
+    FSM,
+    Alphabet,
+    State,
+    TransitionKey,
+    _AnythingElseCls,
+    anything_else,
+)
 
 if TYPE_CHECKING:
     from outlines_core.models.tokenizer import Tokenizer
@@ -479,4 +479,4 @@ def create_fsm_index_tokenizer(
         if subset is not None:
             subset[tokenizer.eos_token_id] = state
 
-    return states_to_token_subsets, empty_token_ids
+    return states_to_token_subsets, empty_token_ids

bindings/python/outlines_core/fsm/types.py

@@ -78,4 +78,4 @@ def datetime_format_fn(sequence: str) -> datetime.datetime:
     else:
         raise NotImplementedError(
             f"The Python type {python_type} is not supported. Please open an issue."
-        )
+        )

bindings/python/outlines_core/integrations/utils.py

@@ -38,4 +38,4 @@ def convert_token_to_string(token: Union[str, bytes]) -> str:
 
     tokenizer.convert_token_to_string = convert_token_to_string
 
-    return tokenizer
+    return tokenizer

bindings/python/outlines_core/models/__init__.py

@@ -10,4 +10,4 @@
 
 from .transformers import Transformers, TransformerTokenizer, mamba, transformers
 
-LogitsGenerator = Union[Transformers]
+LogitsGenerator = Union[Transformers]

bindings/python/outlines_core/models/tokenizer.py

@@ -28,4 +28,4 @@ def convert_token_to_string(self, token: str) -> str:
         represented by the special characted `Ġ`. This prevents matching a raw
         token that includes `Ġ` with a string.
         """
-        ...
+        ...

bindings/python/outlines_core/models/transformers.py

@@ -471,4 +471,4 @@ def mamba(
         model_kwargs=model_kwargs,
         tokenizer_kwargs=tokenizer_kwargs,
         model_class=MambaForCausalLM,
-    )
+    )

bindings/python/src/lib.rs

@@ -1,6 +1,9 @@
-
 use ::outlines_core as core_lib;
-use pyo3::{pyclass, pyfunction, pymethods, pymodule, types::{PyAnyMethods, PyDict, PyModule, PyModuleMethods}, wrap_pyfunction, Bound, PyResult, Python};
+use pyo3::{
+    pyclass, pyfunction, pymethods, pymodule,
+    types::{PyAnyMethods, PyDict, PyModule, PyModuleMethods},
+    wrap_pyfunction, Bound, PyResult, Python,
+};
 use std::collections::{HashMap, HashSet};
 
 #[pymodule]
@@ -131,7 +134,7 @@ pub fn create_fsm_index_end_to_end<'py>(
             &vocabulary_transition_keys,
             start_state,
         );
-        
+
         for (token_id, end_state) in token_ids_end_states {
             if let Ok(existing_dict) = states_to_token_subsets.get_item(start_state) {
                 existing_dict.set_item(token_id, end_state).unwrap();

bindings/python/tests/fsm/test_fsm.py

@@ -88,4 +88,4 @@ def convert_token_to_string(self, token):
     assert fsm.is_final_state(state)
 
     state = fsm.next_state(state=5, token_id=103)
-    assert fsm.is_final_state(state)
+    assert fsm.is_final_state(state)

bindings/python/tests/fsm/test_regex.py

@@ -521,4 +521,4 @@ def test_reduced_vocabulary_with_rare_tokens(rare_token):
     tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2")
     tokenizer = adapt_tokenizer(tokenizer=tokenizer)
     tokenizer.vocabulary[rare_token] = max(tokenizer.vocabulary.values()) + 1
-    reduced_vocabulary(tokenizer)
+    reduced_vocabulary(tokenizer)

outlines-core/src/lib.rs

@@ -1 +1 @@
-pub mod regex;
+pub mod regex;