Remove analyzer.fit

Remove dead code left from PatternNet and PatternAttribution.

examples/introduction_development.ipynb

@@ -240,20 +240,6 @@
  " def __init__(self, model, disable_model_checks=False):\n",
  " pass\n",
  "\n",
- " def fit(self, *args, **kwargs):\n",
- " \"\"\"\n",
- " Train analyzer like a Keras model.\n",
- " Does not need to be implemented.\n",
- " \"\"\"\n",
- " pass\n",
- "\n",
- " def fit_generator(self, *args, **kwargs):\n",
- " \"\"\"\n",
- " Train analyzer like a Keras model.\n",
- " Does not need to be implemented.\n",
- " \"\"\"\n",
- " pass\n",
- "\n",
  " def analyze(self, X):\n",
  " \"\"\"\n",
  " Analyze the behavior of model on input `X`.\n",
@@ -977,7 +963,7 @@
  ],
  "metadata": {
  "kernelspec": {
- "display_name": "Python 3",
+ "display_name": "Python 3.9.7 ('innvestigate-pdNhrmV2-py3.9')",
  "language": "python",
  "name": "python3"
  },
@@ -992,6 +978,11 @@
  "nbconvert_exporter": "python",
  "pygments_lexer": "ipython3",
  "version": "3.9.7"
+ },
+ "vscode": {
+ "interpreter": {
+ "hash": "6d5a31c854ee050015bee53401f829bd4e104bd774bbc6a8c859288cc9ab765e"
+ }
  }
  },
  "nbformat": 4,

src/innvestigate/analyzer/base.py

@@ -31,7 +31,6 @@ class AnalyzerBase(metaclass=ABCMeta):
 
  >>> model = create_keras_model()
  >>> a = Analyzer(model)
- >>> a.fit(X_train) # If analyzer needs training.
  >>> analysis = a.analyze(X_test)
  >>>
  >>> state = a.save()
@@ -127,39 +126,6 @@ def _do_model_checks(self) -> None:
 
  self._model_check_done = True
 
- def fit(self, *_args, disable_no_training_warning: bool = False, **_kwargs):
- """
- Stub that eats arguments. If an analyzer needs training
- include :class:`TrainerMixin`.
-
- :param disable_no_training_warning: Do not warn if this function is
- called despite no training is needed.
- """
- if not disable_no_training_warning:
- # issue warning if no training is foreseen, but fit() is still called.
- warnings.warn(
- "This analyzer does not need to be trained." " Still fit() is called.",
- RuntimeWarning,
- )
-
- def fit_generator(
- self, *_args, disable_no_training_warning: bool = False, **_kwargs
- ):
- """
- Stub that eats arguments. If an analyzer needs training
- include :class:`TrainerMixin`.
-
- :param disable_no_training_warning: Do not warn if this function is
- called despite no training is needed.
- """
- if not disable_no_training_warning:
- # issue warning if no training is foreseen, but fit() is still called.
- warnings.warn(
- "This analyzer does not need to be trained."
- " Still fit_generator() is called.",
- RuntimeWarning,
- )
-
  @abstractmethod
  def analyze(
  self, X: OptionalList[np.ndarray], *args: Any, **kwargs: Any
@@ -248,52 +214,3 @@ def load_npz(fname):
  class_name = npz_file["class_name"].item()
  state = npz_file["state"].item()
  return AnalyzerBase.load(class_name, state)
-
-
-###############################################################################
-
-
-class TrainerMixin:
- """Mixin for analyzer that adapt to data.
-
- This convenience interface exposes a Keras like training routing
- to the user.
- """
-
- # TODO: extend with Y
- def fit(self, X: np.ndarray | None = None, batch_size: int = 32, **kwargs) -> None:
- """
- Takes the same parameters as Keras's :func:`model.fit` function.
- """
- generator = isequence.BatchSequence(X, batch_size)
- return self._fit_generator(generator, **kwargs) # type: ignore
-
- def fit_generator(self, *args, **kwargs):
- """
- Takes the same parameters as Keras's :func:`model.fit_generator`
- function.
- """
- return self._fit_generator(*args, **kwargs)
-
- def _fit_generator(self, *_args, **_kwargs):
- raise NotImplementedError()
-
-
-class OneEpochTrainerMixin(TrainerMixin):
- """Exposes the same interface and functionality as :class:`TrainerMixin`
- except that the training is limited to one epoch.
- """
-
- def fit(self, *args, **kwargs) -> None:
- """
- Same interface as :func:`fit` of :class:`TrainerMixin` except that
- the parameter epoch is fixed to 1.
- """
- return super().fit(*args, epochs=1, **kwargs)
-
- def fit_generator(self, *args, steps: int = None, **kwargs):
- """
- Same interface as :func:`fit_generator` of :class:`TrainerMixin` except that
- the parameter epoch is fixed to 1.
- """
- return super().fit_generator(*args, steps_per_epoch=steps, epochs=1, **kwargs)

tests/dryrun.py

@@ -97,12 +97,10 @@ def _apply_test(self, model: Model) -> None:
  # Generate random training input
  input_shape = model.input_shape[1:]
  x = np.random.rand(1, *input_shape).astype(np.float32)
- x_fit = np.random.rand(16, *input_shape).astype(np.float32)
  # Call model with test input
  model.predict(x)
  # Get analyzer.
  analyzer = self._method(model)
- analyzer.fit(x_fit)
  # Generate random test input
  analysis = analyzer.analyze(x)
  assert tuple(analysis.shape) == (1,) + input_shape