unit8co · gdevos010 · Sep 23, 2024 · Sep 28, 2024 · Sep 29, 2024 · Sep 29, 2024
@@ -21,7 +21,7 @@ repos:
  - id: detect-private-key
 
  - repo: https:/astral-sh/ruff-pre-commit
- rev: v0.3.5
+ rev: v0.6.8
  hooks:
  # try to fix what is possible
  - id: ruff

@@ -25,6 +25,7 @@ but cannot always guarantee backwards compatibility. Changes that may **break co
  - Updated the Ray Tune Hyperparameter Optimization example in the [user guide](https://unit8co.github.io/darts/userguide/hyperparameter_optimization.html) to work with the latest `ray` versions (`>=2.31.0`). [#2459](https:/unit8co/darts/pull/2459) by [He Weilin](https:/cnhwl).
  - Indicate that `multi_models=False` induce a lags shift for each step in `output_chunk_length` in `RegressionModel` and `LinearRegressionModel`. [#2511](https:/unit8co/darts/pull/2511) by [Antoine Madrona](https:/madtoinou).
  - Added reference to `timeseries_generation.datetime_attribute_timeseries` in `TimeSeries.add_datetime_attribute` (0-indexing of encoding is enforced). [#2511](https:/unit8co/darts/pull/2511) by [Antoine Madrona](https:/madtoinou).
+ - Update ruff to v6.8. Also fixed many typos [#2548](https:/unit8co/darts/pull/2548) by [Greg DeVosNouri](https:/gdevos010)
 
 **Fixed**
 
@@ -228,11 +229,11 @@ but cannot always guarantee backwards compatibility. Changes that may **break co
 **Fixed**
 
 - Fixed a bug when using `RegressionModel` with `lags=None`, some `lags_*covariates`, and the covariates starting after or at the same time as the first predictable time step; the lags were not extracted from the correct indices. [#2176](https:/unit8co/darts/pull/2176) by [Dennis Bader](https:/dennisbader).
-- Fixed a bug when calling `window_transform` on a `TimeSeries` with a hierarchy. The hierarchy is now only preseved for single transformations applied to all components, or removed otherwise. [#2207](https:/unit8co/darts/pull/2207) by [Antoine Madrona](https:/madtoinou).
+- Fixed a bug when calling `window_transform` on a `TimeSeries` with a hierarchy. The hierarchy is now only preserved for single transformations applied to all components, or removed otherwise. [#2207](https:/unit8co/darts/pull/2207) by [Antoine Madrona](https:/madtoinou).
 - Fixed a bug in probabilistic `LinearRegressionModel.fit()`, where the `model` attribute was not pointing to all underlying estimators. [#2205](https:/unit8co/darts/pull/2205) by [Antoine Madrona](https:/madtoinou).
 - Raise an error in `RegressionEsembleModel` when the `regression_model` was created with `multi_models=False` (not supported). [#2205](https:/unit8co/darts/pull/2205) by [Antoine Madrona](https:/madtoinou).
 - Fixed a bug in `coefficient_of_variation()` with `intersect=True`, where the coefficient was not computed on the intersection. [#2202](https:/unit8co/darts/pull/2202) by [Antoine Madrona](https:/madtoinou).
-- Fixed a bug in `gridsearch()` with `use_fitted_values=True`, where the model was not propely instantiated for sanity checks. [#2222](https:/unit8co/darts/pull/2222) by [Antoine Madrona](https:/madtoinou).
+- Fixed a bug in `gridsearch()` with `use_fitted_values=True`, where the model was not properly instantiated for sanity checks. [#2222](https:/unit8co/darts/pull/2222) by [Antoine Madrona](https:/madtoinou).
 - Fixed a bug in `TimeSeries.append/prepend_values()`, where the component names and the hierarchy were dropped. [#2237](https:/unit8co/darts/pull/2237) by [Antoine Madrona](https:/madtoinou).
 - Fixed a bug in `get_multioutput_estimator()`, where the index of the estimator was incorrectly calculated. [#2246](https:/unit8co/darts/pull/2246) by [Antoine Madrona](https:/madtoinou).
 - 🔴 Fixed a bug in `datetime_attribute_timeseries()`, where 1-indexed attributes were not properly handled. Also, 0-indexing is now enforced for all the generated encodings. [#2242](https:/unit8co/darts/pull/2242) by [Antoine Madrona](https:/madtoinou).

@@ -165,7 +165,7 @@ def fit_detect(
  Returns
  -------
  Union[TimeSeries, Sequence[TimeSeries]]
- Binary prediciton (1 if considered as an anomaly, 0 if not)
+ Binary prediction (1 if considered as an anomaly, 0 if not)
  """
  self.fit(series)
  return self.detect(series, name="series")

@@ -666,7 +666,7 @@ def _plot_series(series, ax_id, linewidth, label_name, **kwargs):
  series
  The series to plot.
  ax_id
- The axis the series will be ploted on.
+ The axis the series will be plotted on.
  linewidth
  Thickness of the line.
  label_name

@@ -54,14 +54,14 @@ def __init__(
  Whether to remove the NaNs from the start and the end of the transformed series.
  drop_static_covariates
  If set to `True`, the statics covariates of the input series won't be transferred to the output.
- This migth be useful for multivariate series with component-specific static covariates.
+ This might be useful for multivariate series with component-specific static covariates.
  name
  A specific name for the scaler
  n_jobs
  The number of jobs to run in parallel. Parallel jobs are created only when a ``Sequence[TimeSeries]`` is
- passed as input to a method, parallelising operations regarding different ``TimeSeries``. Defaults to `1`
+ passed as input to a method, parallelizing operations regarding different ``TimeSeries``. Defaults to `1`
  (sequential). Setting the parameter to `-1` means using all the available processors.
- Note: for a small amount of data, the parallelisation overhead could end up increasing the total
+ Note: for a small amount of data, the parallelization overhead could end up increasing the total
  required amount of time.
  verbose
  Optionally, whether to print operations progress

@@ -260,9 +260,9 @@ def _create_component_masks(
  """
  Returns a boolean array indicating which components of the UNTRANSFORMED
  `stat_covs` are numerical and a boolean array indicating which components
- of the UNTRANSFORMED `stat_covs` are categoical.
+ of the UNTRANSFORMED `stat_covs` are categorical.
 
- It's important to recognise that these masks only apply to the UNTRANSFORMED
+ It's important to recognize that these masks only apply to the UNTRANSFORMED
  static covariates since some transformations can generate multiple new components
  from a single component (e.g. one-hot encoding).
  """
@@ -327,9 +327,9 @@ def _create_inv_component_masks(
  """
  Returns a boolean array indicating which components of the TRANSFORMED
  `stat_covs` are numerical and a boolean array indicating which components
- of the TRANSFORMED `stat_covs` are categoical.
+ of the TRANSFORMED `stat_covs` are categorical.
 
- It's important to recognise that these masks only apply to the UNTRANSFORMED
+ It's important to recognize that these masks only apply to the UNTRANSFORMED
  static covariates since some transformations can generate multiple new components
  from a single component (e.g. one-hot encoding).
  """

@@ -398,7 +398,7 @@ def summary_plot(
  Returns
  -------
  shaps_
- A nested dictionary {horizon : {component : shap.Explaination}} containing the raw Explanations for all
+ A nested dictionary {horizon : {component : shap.Explanation}} containing the raw Explanations for all
  the horizons and components.
  """
 

@@ -3,7 +3,7 @@
 -------
 
 For deterministic forecasts (point predictions with `num_samples == 1`), probabilistic forecasts (`num_samples > 1`),
-and quantile forecasts. For probablistic and quantile forecasts, use parameter `q` to define the quantile(s) to
+and quantile forecasts. For probabilistic and quantile forecasts, use parameter `q` to define the quantile(s) to
 compute the deterministic metrics on:
 
  - Aggregated over time:

@@ -287,14 +287,14 @@ def wrapper_multivariate_support(*args, **kwargs) -> METRIC_OUTPUT_TYPE:
  raise_log(
  ValueError(
  "`q` must be of tuple of `(np.ndarray, Optional[pd.Index])` "
- "where the (quantile values, optioanl quantile component names). "
+ "where the (quantile values, optional quantile component names). "
  f"Received `q={q}`."
  ),
  logger=logger,
  )
  q, q_comp_names = q
  if not pred_series.is_stochastic:
- # quantile component names are required if the predictions are not stochastic (as for stocahstic
+ # quantile component names are required if the predictions are not stochastic (as for stochastic
  # predictions, the quantiles can be retrieved from the sample dimension for each component)
  if q_comp_names is None:
  q_comp_names = pd.Index(
@@ -611,7 +611,7 @@ def _get_error_scale(
  """Computes the error scale based on a naive seasonal forecasts on `insample` values with seasonality `m`."""
  if not isinstance(m, int):
  raise_log(
- ValueError(f"Seasonality `m` must be of type `int`, recevied `m={m}`"),
+ ValueError(f"Seasonality `m` must be of type `int`, received `m={m}`"),
  logger=logger,
  )
 

@@ -257,7 +257,7 @@ def __init__(
  hidden_fc_sizes
  Sizes of hidden layers connecting the last hidden layer of the RNN module to the output, if any.
  dropout
- Fraction of neurons afected by Dropout.
+ Fraction of neurons affected by Dropout.
  activation
  The name of a torch.nn activation function to be applied between the layers of the fully connected network.
  Default: "ReLU".

@@ -3144,7 +3144,7 @@ def _verify_passed_predict_covariates(self, future_covariates):
 
  @property
  def _supress_generate_predict_encoding(self) -> bool:
- """Controls wether encodings should be generated in :func:`FutureCovariatesLocalForecastingModel.predict()``"""
+ """Controls whether encodings should be generated in :func:`FutureCovariatesLocalForecastingModel.predict()``"""
  return False
 
  @property

@@ -115,7 +115,7 @@ def __init__(
  To add a new naive model:
  - subclass from `_GlobalNaiveModel` with implementation of private method `_create_model` that creates an
  object of:
- - subclass from `_GlobalNaiveModule` with implemention of private method `_forward`
+ - subclass from `_GlobalNaiveModule` with implementation of private method `_forward`
 
  .. note::
  - Model checkpointing with `save_checkpoints=True`, and checkpoint loading with `load_from_checkpoint()`

@@ -251,7 +251,7 @@ def fit(
 
  self._model_container[quantile] = self.model
 
- # replace the last trained QuantileRegressor with the dictionnary of Regressors.
+ # replace the last trained QuantileRegressor with the dictionary of Regressors.
  self.model = self._model_container
 
  return self

@@ -529,7 +529,7 @@ def _produce_predict_output(self, x: Tuple) -> torch.Tensor:
  def on_save_checkpoint(self, checkpoint: Dict[str, Any]) -> None:
  # we must save the dtype for correct parameter precision at loading time
  checkpoint["model_dtype"] = self.dtype
- # we must save the shape of the input to be able to instanciate the model without calling fit_from_dataset
+ # we must save the shape of the input to be able to instantiate the model without calling fit_from_dataset
  checkpoint["train_sample_shape"] = self.train_sample_shape
  # we must save the loss to properly restore it when resuming training
  checkpoint["loss_fn"] = self.criterion

@@ -111,7 +111,7 @@ def encode_year(idx):
  }
  ..
  cap
- Parameter specifiying the maximum carrying capacity when predicting with logistic growth.
+ Parameter specifying the maximum carrying capacity when predicting with logistic growth.
  Mandatory when `growth = 'logistic'`, otherwise ignored.
  See <https://facebook.github.io/prophet/docs/saturating_forecasts.html> for more information
  on logistic forecasts.
@@ -121,7 +121,7 @@ def encode_year(idx):
  - a function taking a DatetimeIndex or RangeIndex and returning a corresponding a Sequence of numbers,
  where each number indicates the carrying capacity at this index.
  floor
- Parameter specifiying the minimum carrying capacity when predicting logistic growth.
+ Parameter specifying the minimum carrying capacity when predicting logistic growth.
  Optional when `growth = 'logistic'` (defaults to 0), otherwise ignored.
  See <https://facebook.github.io/prophet/docs/saturating_forecasts.html> for more information
  on logistic forecasts.

@@ -458,12 +458,12 @@ def ensemble(
  ensembled = [
  self.regression_model.predict(
  n=len(prediction),
- series=serie,
+ series=series,
  future_covariates=prediction,
  num_samples=num_samples,
  predict_likelihood_parameters=predict_likelihood_parameters,
  )
- for serie, prediction in zip(series, predictions)
+ for series, prediction in zip(series, predictions)
  ]
  return seq2series(ensembled) if is_single_series else ensembled
 

@@ -369,7 +369,7 @@ def _generate_lags(
  raise_log(
  ValueError(
  f"`{lags_name}` - `{comp_name}`: must be either a {supported_types}. "
- f"Gived : {type(comp_lags)}."
+ f"Given : {type(comp_lags)}."
  ),
  logger,
  )

@@ -318,7 +318,7 @@ def __init__(
  n_rnn_layers
  The number of recurrent layers.
  dropout
- Fraction of neurons afected by Dropout.
+ Fraction of neurons affected by Dropout.
  training_length
  The length of both input (target and covariates) and output (target) time series used during
  training. Must have a larger value than `input_chunk_length`, because otherwise during training

@@ -292,7 +292,7 @@ def test_aggregator_predict_wrong_inputs(self, config):
 
  @pytest.mark.parametrize("config", list_NonFittableAggregator)
  def test_NonFittableAggregator_predict(self, config):
- """Check that predict() works as intented"""
+ """Check that predict() works as intended"""
  aggregator_cls, cls_kwargs, _ = config
  aggregator = aggregator_cls(**cls_kwargs)
 

@@ -1262,7 +1262,7 @@ def test_PyODScorer(self):
  with pytest.raises(ValueError):
  PyODScorer(model=KNN(), window=0)
 
- # diff_fn paramter
+ # diff_fn parameter
  # must be None, 'diff' or 'abs_diff'
  with pytest.raises(ValueError):
  PyODScorer(model=KNN(), diff_fn="random")
@@ -1531,7 +1531,7 @@ def helper_evaluate_nll_scorer(
  np.array(deterministic_values).reshape(2, 2, -1)
  )
 
- # compute the NLL values witn score_from_prediction for scorer with window=1 and 2
+ # compute the NLL values with score_from_prediction for scorer with window=1 and 2
  # t -> timestamp, c -> component and w -> window used in scorer
  value_t1_c1_w1 = NLLscorer_w1.score_from_prediction(
  series[0]["0"], distribution_series[0]["0"]
@@ -1570,7 +1570,7 @@ def helper_evaluate_nll_scorer(
  )
 
  # multivariate case
- # compute the NLL values witn score_from_prediction for scorer with window=1 and window=2
+ # compute the NLL values with score_from_prediction for scorer with window=1 and window=2
  value_t1_2_c1_2_w1 = NLLscorer_w1.score_from_prediction(
  series, distribution_series
  )
@@ -1686,7 +1686,7 @@ def test_window_greater_than_one(self, window, model, series):
  assert score_T[-1] == score_F[-1]
 
  def test_fun_window_agg(self):
- """Verify that the anomaly score aggregation works as intented"""
+ """Verify that the anomaly score aggregation works as intended"""
  # window = 2, alternating anomaly scores
  window = 2
  scorer = KMeansScorer(window=window)

@@ -540,8 +540,8 @@ def test_include_current(self):
  )
  def test_ts_windowtransf_hierarchy(self, transforms):
  """Checking that supported transforms behave as expected:
- - implicitely applied to all components
- - passing explicitely all components
+ - implicitly applied to all components
+ - passing explicitly all components
  """
  ts = helper_generate_ts_hierarchy(10)
 

@@ -37,7 +37,7 @@ def test_rin(self):
  rin = RINorm(input_dim=7, affine=affine)
  x_norm = rin(x)
 
- # expand dims to simulate probablistic forecasting
+ # expand dims to simulate probabilistic forecasting
  x_denorm = rin.inverse(x_norm.view(x_norm.shape + (1,))).squeeze(-1)
  assert torch.all(torch.isclose(x, x_denorm)).item()
 

@@ -1122,15 +1122,15 @@ def test_gridsearch_sample_weight(self, config):
  else:
  sample_weight = "linear"
 
- paramameters = {"lags": [3], "output_chunk_length": [1]}
+ parameters = {"lags": [3], "output_chunk_length": [1]}
  start_kwargs = {"start": -1, "start_format": "position"}
  gs_kwargs = {"val_series": ts} if use_val_series else {"forecast_horizon": 1}
  gs_non_weighted = LinearRegressionModel.gridsearch(
- paramameters, series=ts[:-1], **start_kwargs, **gs_kwargs
+ parameters, series=ts[:-1], **start_kwargs, **gs_kwargs
  )[-1]
 
  gs_weighted = LinearRegressionModel.gridsearch(
- paramameters,
+ parameters,
  series=ts[:-1],
  sample_weight=sample_weight,
  **start_kwargs,

@@ -82,7 +82,7 @@ def test_trained_models(self):
 
  # both global trained, retrain = True
  with pytest.raises(ValueError):
- # models need to be explicitely reset before retraining them
+ # models need to be explicitly reset before retraining them
  NaiveEnsembleModel(
  [global_model, global_model], train_forecasting_models=True
  )

@@ -2323,7 +2323,7 @@ def test_predict_likelihood_parameters(self, model_type):
  ),
  )
  def test_probabilistic_optimized_hist_fc_regression(self, config):
- """Tests optimized probilistic historical forecasts for regression models."""
+ """Tests optimized probabilistic historical forecasts for regression models."""
  np.random.seed(42)
  lpo, multi_models, n = config
  ocl = 2

@@ -520,7 +520,7 @@ def test_predict_likelihood_parameters_univariate_torch_models(
  n_samples,
  ))
 
- # Dirichlet must be handled sligthly differently since its multivariate
+ # Dirichlet must be handled slightly differently since its multivariate
  if isinstance(lkl, DirichletLikelihood):
  values = torch.swapaxes(values, 1, 3)
  values = torch.squeeze(values, 3)

@@ -121,7 +121,7 @@ def test_builtin_extended_trainer(self):
  )
  model.fit(self.series, epochs=1)
 
- # flaot 16 not supported
+ # float 16 not supported
  with pytest.raises(ValueError):
  invalid_trainer_kwarg = {
  "precision": "16-mixed",