Merge pull request #8 from neulab/neubig/fix_continuous_integration

Fixed continuous integration

.github/workflows/ci.yml

@@ -1,5 +1,6 @@
+---
 name: Continuous Integration
-on: # yamllint disable-line rule:truthy
+on:  # yamllint disable-line rule:truthy
   push:
     branches:
       - main
@@ -18,7 +19,7 @@ jobs:
       - name: Install ruff
         run: pip install ruff
       - name: Run ruff
-        run: ruff --check .
+        run: ruff check .
   python-mypy:
     runs-on: ubuntu-latest
     steps:
@@ -31,14 +32,6 @@ jobs:
         uses: pre-commit/action@v3.0.0
         with:
           extra_args: mypy --all-files
-  lint-requirements-txt:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v3
-      - name: lint
-        uses: pre-commit/action@v3.0.0
-        with:
-          extra_args: requirements-txt-fixer --all-files
   lint-markdown:
     runs-on: ubuntu-latest
     steps:
@@ -75,19 +68,3 @@ jobs:
         uses: pre-commit/action@v3.0.0
         with:
           extra_args: shellcheck --all-files
-  lint-trailing-whitespace:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v3
-      - name: lint
-        uses: pre-commit/action@v3.0.0
-        with:
-          extra_args: trailing-whitespace --all-files
-  lint-eof-newline:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v3
-      - name: lint
-        uses: pre-commit/action@v3.0.0
-        with:
-          extra_args: end-of-file-fixer --all-files

.github/workflows/release.yml

@@ -1,10 +1,9 @@
+---
 name: Release workflow
-
-on:
+on:  # yamllint disable-line rule:truthy
   push:
     tags:
       - "v[0123456789].*"
-
 jobs:
   release:
     runs-on: ubuntu-latest

.gitignore

@@ -2,4 +2,4 @@ __pycache__
 build
 llments.egg-info
 .vscode
-.pytest_cache
+.pytest_cache

.pre-commit-config.yaml

@@ -1,19 +1,20 @@
+---
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.1.5
+    rev: v0.2.1
     hooks:
       - id: ruff
       - id: ruff-format
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: 'v1.6.1'
+    rev: v1.8.0
     hooks:
-    -   id: mypy
+      - id: mypy
   - repo: https://github.com/DavidAnson/markdownlint-cli2
-    rev: v0.10.0
+    rev: v0.12.1
     hooks:
       - id: markdownlint-cli2
   - repo: https://github.com/adrienverge/yamllint
-    rev: v1.33.0
+    rev: v1.34.0
     hooks:
       - id: yamllint
   - repo: https://github.com/koalaman/shellcheck-precommit

README.md

@@ -2,7 +2,17 @@
 
 ## Development Information
 
+To start developing, install the development dependencies and pre-commit hooks:
+
 ```bash
-pip install .
+pip install ".[dev]"
 pre-commit install
 ```
+
+There are several [pre-commit hooks](https://pre-commit.com/) that will run on every
+commit to perform formatting, typing, and linting.
+
+* `ruff` - Runs formatting, import sorting, and linting.
+* `mypy` - Runs type checking.
+* `markdownlint` - Runs markdown linting.
+* `yamllint` - Runs YAML linting.

llments/distance/distance.py

@@ -1,12 +1,10 @@
-
 import abc
 
 from llments.lm.lm import LanguageModel
 
 
 class Distance:
-
-    @abc.abstractclassmethod
+    @abc.abstractmethod
     def distance(self, lm1: LanguageModel, lm2: LanguageModel) -> float:
         """Returns a distance between two language models."""
         ...

llments/distance/norm.py

@@ -1,4 +1,3 @@
-
 from llments.lm.lm import LanguageModel
 
 

llments/eval/eval.py

@@ -1,12 +1,11 @@
-
 import abc
 import dataclasses
 
 
 class PairwiseEvaluator:
     """A class that defines an evaluation function, assessing a hypothesized string."""
 
-    @abc.abstractclassmethod
+    @abc.abstractmethod
     def evaluate(self, hyp: str, ref: str) -> float:
         """Returns an evaluation score between 0 and 1 for two strings.
 
@@ -25,7 +24,7 @@ class EvaluatorMetadata:
 class GeneralEvaluator:
     """A class that defines an evaluation function, assessing a hypothesized string."""
 
-    @abc.abstractclassmethod
+    @abc.abstractmethod
     def evaluate(self, hyp: str, ref: EvaluatorMetadata) -> float:
         """Returns an evaluation score between 0 and 1 for two strings.
 

llments/lm/empirical.py

@@ -1,3 +1,4 @@
+from typing import Any
 from llments.lm.lm import LanguageModel
 import random
 import json
@@ -10,10 +11,10 @@ def __init__(self, data: list[str], probs: list[float] | None = None):
             probs = [1 / len(data)] * len(data)
         self.data = pd.DataFrame({"text": data, "prob": probs})
 
-    def sample(self, condition: str | None):
+    def generate(self, condition: str | None, **kwargs: Any) -> str:
         """Sample from the language model, possibly conditioned on a prefix."""
         if condition is None:
-            return random.choice(self.data["text"], p=self.data["probs"])
+            return random.choices(self.data["text"], weights=self.data["probs"])[0]
         else:
             # Filter to only those that start with the condition
             filtered_df = self.data[self.data["text"].str.startswith(condition)]
@@ -24,9 +25,9 @@ def sample(self, condition: str | None):
                 )
             # Normalize the probabilities
             filtered_df["prob"] = filtered_df["prob"] / filtered_df["prob"].sum()
-            return random.choice(filtered_df["text"], p=filtered_df["probs"])
+            return random.choices(filtered_df["text"], weights=filtered_df["probs"])[0]
 
-    def fit(self, target: LanguageModel):
+    def fit(self, target: LanguageModel, task_description: str | None = None):
         raise ValueError(
             "Cannot fit an empirical distribution to another distribution."
         )

llments/lm/lm.py

@@ -2,7 +2,7 @@
 
 
 class LanguageModel:
-    @abc.abstractclassmethod
+    @abc.abstractmethod
     def generate(
         self,
         condition: str | None,
@@ -11,7 +11,7 @@ def generate(
         """Generate from the language model, possibly conditioned on a prefix."""
         ...
 
-    @abc.abstractclassmethod
+    @abc.abstractmethod
     def fit(
         self, target: "LanguageModel", task_description: str | None = None
     ) -> "LanguageModel":

llments/lm/lm_loader.py

@@ -1,5 +1,3 @@
-
-
 from llments.lm.lm import LanguageModel
 
 

llments/lm/lm_loader_test.py

@@ -1,4 +1,3 @@
-
 def test_load_from_spec_file():
     """Test that load_from_specification_file() loads a language model."""
     raise NotImplementedError