Replace black with ruff and add pre-commit

.github/workflows/style.yml

@@ -16,10 +16,10 @@ jobs:
         uses: actions/setup-python@v2
         with:
           python-version: ${{ matrix.python-version }}
-      - name: Install Black
+      - name: Install Ruff
         run: |
           python -m pip install --upgrade pip
-          pip install black
+          pip install ruff
       - name: Check style
         run: |
-          black --check *.py deepsensor tests
+          ruff check

.pre-commit-config.yaml

@@ -0,0 +1,9 @@
+repos:
+- repo: https://github.com/astral-sh/ruff-pre-commit
+  # Ruff version.
+  rev: v0.4.2
+  hooks:
+    # Run the linter.
+    - id: ruff
+    # Run the formatter.
+    - id: ruff-format

CONTRIBUTING.md

@@ -95,6 +95,20 @@ That is, please do not use the [rebase](https://help.github.com/en/articles/abou
 command to edit previous commit messages, combine multiple commits into one, or delete or revert
 commits that are no longer necessary.
 
+Make sure you're using the developer dependencies.
+If you're working locally on the source code, *before* commiting, please run `pip install -r requirements/requirements.dev.txt` to install some useful dependencies just for development.
+This includes `pre-commit` and `ruff` which are used to check and format your code style when you run `git commit`, so that you don't have to.
+
+Using pre-commit:
++ To make this work, just run `pre-commit install` and when you commit, `ruff` will be run to check the style of the files that you've changed.
++ Note, you may find that if `ruff` needs to edit your files, you'll have to run `git add ...` and `git commit ..` again as the pre-commit hook will stop the commit until the changes pass its tests, `ruff` will also have slightly edited the files you added, so you'll need to stage and commit again.
+
+Without pre-commit:
++ Alternatively, you can run `ruff` yourself (without) `pre-commit` by installing `ruff` as above and just running `ruff format`.
+
+You should also run `pytest` and check that your changes don't break any of the existing tests.
+If you've made changes to the source code, you may need to add some tests to make sure that they don't get broken in the future.
+
 #### 4. Open a Pull Request
 
 We encourage you to open a pull request as early in your contributing process as possible.

deepsensor/__init__.py

@@ -1,5 +1,5 @@
 class Backend:
-    """Backend for deepsensor
+    """Backend for deepsensor.
 
     This class is used to provide a consistent interface for either tensorflow or
     pytorch backends. It is used to assign the backend to the deepsensor module.

deepsensor/active_learning/acquisition_fns.py

@@ -10,9 +10,7 @@
 
 
 class AcquisitionFunction:
-    """
-    Parent class for acquisition functions.
-    """
+    """Parent class for acquisition functions."""
 
     # Class attribute to indicate whether the acquisition function should be
     #   minimised or maximised
@@ -24,23 +22,21 @@ def __init__(
         context_set_idx: int = 0,
         target_set_idx: int = 0,
     ):
-        """
-        Args:
-            model (:class:`~.model.model.ProbabilisticModel`):
-                [Description of the model parameter.]
-            context_set_idx (int):
-                Index of context set to add new observations to when computing
-                the acquisition function.
-            target_set_idx (int):
-                Index of target set to compute acquisition function for.
+        """Args:
+        model (:class:`~.model.model.ProbabilisticModel`):
+            [Description of the model parameter.]
+        context_set_idx (int):
+            Index of context set to add new observations to when computing
+            the acquisition function.
+        target_set_idx (int):
+            Index of target set to compute acquisition function for.
         """
         self.model = model
         self.context_set_idx = context_set_idx
         self.target_set_idx = target_set_idx
 
     def __call__(self, task: Task, *args, **kwargs) -> np.ndarray:
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -61,21 +57,18 @@ def __call__(self, task: Task, *args, **kwargs) -> np.ndarray:
 
 
 class AcquisitionFunctionOracle(AcquisitionFunction):
-    """
-    Signifies that the acquisition function is computed using the true
+    """Signifies that the acquisition function is computed using the true
     target values.
     """
 
 
 class AcquisitionFunctionParallel(AcquisitionFunction):
-    """
-    Parent class for acquisition functions that are computed across all search
+    """Parent class for acquisition functions that are computed across all search
     points in parallel.
     """
 
     def __call__(self, task: Task, X_s: np.ndarray, **kwargs) -> np.ndarray:
-        """
-        ...
+        """...
 
         :param **kwargs:
         :no-index:
@@ -104,8 +97,7 @@ class MeanStddev(AcquisitionFunction):
     min_or_max = "min"
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -126,8 +118,7 @@ class MeanVariance(AcquisitionFunction):
     min_or_max = "min"
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -148,8 +139,7 @@ class pNormStddev(AcquisitionFunction):
     min_or_max = "min"
 
     def __init__(self, *args, p: int = 1, **kwargs):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -161,8 +151,7 @@ def __init__(self, *args, p: int = 1, **kwargs):
         self.p = p
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -185,8 +174,7 @@ class MeanMarginalEntropy(AcquisitionFunction):
     min_or_max = "min"
 
     def __call__(self, task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -208,8 +196,7 @@ class JointEntropy(AcquisitionFunction):
     min_or_max = "min"
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -230,8 +217,7 @@ class OracleMAE(AcquisitionFunctionOracle):
     min_or_max = "min"
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -256,8 +242,7 @@ class OracleRMSE(AcquisitionFunctionOracle):
     min_or_max = "min"
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -282,8 +267,7 @@ class OracleMarginalNLL(AcquisitionFunctionOracle):
     min_or_max = "min"
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -308,8 +292,7 @@ class OracleJointNLL(AcquisitionFunctionOracle):
     min_or_max = "min"
 
     def __call__(self, task: Task):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -330,8 +313,7 @@ class Random(AcquisitionFunctionParallel):
     min_or_max = "max"
 
     def __init__(self, *args, seed: int = 42, **kwargs):
-        """
-        ...
+        """...
 
         :no-index:
 
@@ -343,8 +325,7 @@ def __init__(self, *args, seed: int = 42, **kwargs):
         self.rng = np.random.default_rng(seed)
 
     def __call__(self, task: Task, X_s: np.ndarray, **kwargs):
-        """
-        ...
+        """...
 
         :param **kwargs:
         :no-index:
@@ -368,8 +349,7 @@ class ContextDist(AcquisitionFunctionParallel):
     min_or_max = "max"
 
     def __call__(self, task: Task, X_s: np.ndarray, **kwargs):
-        """
-        ...
+        """...
 
         :param **kwargs:
         :no-index:
@@ -410,8 +390,7 @@ class Stddev(AcquisitionFunctionParallel):
     min_or_max = "max"
 
     def __call__(self, task: Task, X_s: np.ndarray, **kwargs):
-        """
-        ...
+        """...
 
         :param **kwargs:
         :no-index:
@@ -434,8 +413,7 @@ def __call__(self, task: Task, X_s: np.ndarray, **kwargs):
 
 
 class ExpectedImprovement(AcquisitionFunctionParallel):
-    """
-    Expected improvement acquisition function.
+    """Expected improvement acquisition function.
 
     .. note::
 
@@ -446,8 +424,7 @@ class ExpectedImprovement(AcquisitionFunctionParallel):
     min_or_max = "max"
 
     def __call__(self, task: Task, X_s: np.ndarray, **kwargs) -> np.ndarray:
-        """
-        :param **kwargs:
+        """:param **kwargs:
         :no-index:
 
         Args:

deepsensor/active_learning/algorithms.py

@@ -240,7 +240,7 @@ def _validate_n_new_context(
             )
 
     def _get_times_from_tasks(self):
-        """Get times from tasks"""
+        """Get times from tasks."""
         times = [task["time"] for task in self.tasks]
         # Check for any repeats
         if len(times) != len(set(times)):
@@ -285,7 +285,7 @@ def _model_infill_at_search_points(
         return infill_ds
 
     def _sample_y_infill(self, infill_ds, time, x1, x2):
-        """Sample infill values at a single location"""
+        """Sample infill values at a single location."""
         assert isinstance(infill_ds, (xr.Dataset, xr.DataArray))
         y = infill_ds.sel(time=time, x1=x1, x2=x2)
         if isinstance(y, xr.Dataset):
@@ -298,9 +298,8 @@ def _sample_y_infill(self, infill_ds, time, x1, x2):
             return y
 
     def _build_acquisition_fn_ds(self, X_s: Union[xr.Dataset, xr.DataArray]):
-        """
-        Initialise xr.DataArray for storing acquisition function values on
-        search grid
+        """Initialise xr.DataArray for storing acquisition function values on
+        search grid.
         """
         prepend_dims = ["iteration"]  # , "sample"]  # MC sample TODO
         prepend_coords = {
@@ -320,7 +319,7 @@ def _build_acquisition_fn_ds(self, X_s: Union[xr.Dataset, xr.DataArray]):
         return acquisition_fn_ds
 
     def _init_acquisition_fn_object(self, X_s: xr.Dataset):
-        """Instantiate acquisition function object"""
+        """Instantiate acquisition function object."""
         # Unnormalise before instantiating
         X_s = self.model.data_processor.map_coords(X_s, unnorm=True)
         if isinstance(X_s, (xr.Dataset, xr.DataArray)):
@@ -335,8 +334,7 @@ def _init_acquisition_fn_object(self, X_s: xr.Dataset):
             raise TypeError(f"Unsupported type for X_s: {type(X_s)}")
 
     def _search(self, acquisition_fn: AcquisitionFunction):
-        """
-        Run one greedy pass by looping over each point in ``X_s`` and
+        """Run one greedy pass by looping over each point in ``X_s`` and
         computing the acquisition function.
         """
         importances_list = []
@@ -434,8 +432,7 @@ def _select_best(self, importances, X_s_arr):
         return best_x_query
 
     def _single_greedy_iteration(self, acquisition_fn: AcquisitionFunction):
-        """
-        Run a single greedy grid search iteration and append the optimal
+        """Run a single greedy grid search iteration and append the optimal
         context location to self.X_new.
         """
         importances = self._search(acquisition_fn)
@@ -451,8 +448,7 @@ def __call__(
         tasks: Union[List[Task], Task],
         diff: bool = False,
     ) -> Tuple[pd.DataFrame, xr.Dataset]:
-        """
-        Iteratively propose new context points using the greedy sensor placement algorithm.
+        """Iteratively propose new context points using the greedy sensor placement algorithm.
 
         Args:
             acquisition_fn (:class:`~.active_learning.acquisition_fns.AcquisitionFunction`):

deepsensor/config.py

@@ -1,6 +1,4 @@
-"""
-Configuration file for deepsensor
-"""
+"""Configuration file for deepsensor."""
 
 DEFAULT_LAB_EPSILON = 1e-6
 """