Fixing many issues in the docs (#2343)

deepchecks/tabular/checks/model_evaluation/regression_error_distribution.py

@@ -127,6 +127,7 @@ def add_condition_kurtosis_greater_than(self, threshold: float = -0.1):
 
         Kurtosis is a measure of the shape of the distribution, helping us understand if the distribution
         is significantly "wider" from a normal distribution. A lower value indicates a "wider" distribution.
+
         Parameters
         ----------
         threshold : float , default: -0.1

deepchecks/tabular/checks/model_evaluation/weak_segments_performance.py

@@ -42,6 +42,7 @@ class WeakSegmentsPerformance(SingleDatasetCheck, WeakSegmentAbstract):
     In order to achieve this, the check trains several simple tree based models which try to predict the error of the
     user provided model on the dataset. The relevant segments are detected by analyzing the different
     leafs of the trained trees.
+
     Parameters
     ----------
     columns : Union[Hashable, List[Hashable]] , default: None

deepchecks/tabular/suites/default_suites.py

@@ -74,7 +74,7 @@ def data_integrity(columns: Union[Hashable, List[Hashable]] = None,
            * - :ref:`plot_tabular_identifier_label_correlation`
              - :class:`~deepchecks.tabular.checks.data_integrity.IdentifierLabelCorrelation`
            * - :ref:`plot_tabular_feature_feature_correlation`
-                - :class:`~deepchecks.tabular.checks.data_integrity.FeatureFeatureCorrelation`
+             - :class:`~deepchecks.tabular.checks.data_integrity.FeatureFeatureCorrelation`
 
     Parameters
     ----------
@@ -270,7 +270,7 @@ def model_evaluation(alternative_scorers: Dict[str, Callable] = None,
            * - :ref:`plot_tabular_model_inference_time`
              - :class:`~deepchecks.tabular.checks.model_evaluation.ModelInferenceTime`
            * - :ref:`plot_tabular_train_test_prediction_drift`
-                - :class:`~deepchecks.tabular.checks.model_evaluation.TrainTestPredictionDrift`
+             - :class:`~deepchecks.tabular.checks.model_evaluation.TrainTestPredictionDrift`
 
     Parameters
     ----------

docs/source/checks/vision/model_evaluation/plot_train_test_prediction_drift.py

@@ -74,7 +74,7 @@
 
 #%%
 # Load Dataset
-# ----------------
+# ------------
 
 
 train_ds = load_dataset(train=True, batch_size=64, object_type='VisionData')
@@ -83,7 +83,7 @@
 
 #%%
 # Running TrainTestPredictionDrift on classification
-# ---------------------------------------------
+# --------------------------------------------------
 
 check = TrainTestPredictionDrift()
 result = check.run(train_ds, test_ds)
@@ -99,6 +99,7 @@
 #%%
 # Understanding the results
 # -------------------------
+#
 # We can see there is almost no drift between the train & test predictions. This means the
 # split to train and test was good (as it is balanced and random). Let's check the
 # performance of a simple model trained on MNIST.
@@ -109,7 +110,8 @@
 
 #%%
 # MNIST with prediction drift
-# ======================
+# ===========================
+#
 # Now, let's try to separate the MNIST dataset in a different manner that will result
 # in a prediction drift, and see how it affects the performance. We are going to create a
 # custom `collate_fn`` in the test dataset, that will select a few of the samples with class 0

docs/source/checks/vision/model_evaluation/plot_weak_segments_performance.py

@@ -64,8 +64,10 @@
 # To display the results in an IDE like PyCharm, you can use the following code:
 
 #  result.show_in_window()
+
 #%%
 # The result will be displayed in a new window.
+
 #%%
 # Observe the check's output
 # --------------------------

docs/source/nitpick-exceptions

@@ -18,3 +18,5 @@ py:meth torch.nn.Module.train
 py:func load_state_dict
 py:attr state_dict
 py:meth load_state_dict
+py:meth ignite.metrics
+py:attr ignite.metrics

docs/source/user-guide/general/ci_cd.rst

@@ -59,9 +59,9 @@ Deepchecks can be used in the CI/CD process at 2 main steps of model development
 In this guide we will show end to end examples of validating both the data and the trained model. In most use cases those
 processes will be separated into two separate pipelines, one for data validation and one for model validation.
 We will use the default suites provided by deepchecks, but it's possible to create a
-:doc:`custom suite</user-guide/general/customizations/plot_create_a_custom_suite>`
+:doc:`custom suite</user-guide/general/customizations/examples/plot_create_a_custom_suite>`
 containing hand chosen checks and
-:doc:`conditions</user-guide/general/customizations/plot_configure_check_conditions>`
+:doc:`conditions</user-guide/general/customizations/examples/plot_configure_check_conditions>`
 in order to cater to the specific needs of the project.
 
 Airflow Integration

docs/source/user-guide/general/drift_guide.rst

@@ -118,14 +118,14 @@ In general, it is recommended to use Cramer's V, unless your variable includes c
 However, in cases of a variable with many categories with few samples, it is still recommended to use Cramer's V, as PSI will not be able to detect change in the smaller categories.
 
 Detecting Drift in Unbalanced Classification Tasks
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 In classification problems, it is common to have unbalanced data, meaning that the number of samples in each class is
 highly skewed. For example, in a dataset of credit card transactions, the number of fraudulent transactions is usually
 much lower than the number of non-fraudulent transactions, and can be below 1% of the total number of samples.
 
 In such cases, running the :doc:`TrainTestLabelDrift </checks_gallery/tabular/train_test_validation/plot_train_test_label_drift>`:
-or :doc:`TrainTestPredictionDrift </checks_gallery/tabular/train_test_validation/plot_train_test_prediction_drift>` checks
+or :doc:`TrainTestPredictionDrift </checks_gallery/tabular/model_evaluation/plot_train_test_prediction_drift>` checks
 with the default parameters will likely lead to a false negative, as for example a change in the percent of fraudulent
 transactions from 0.2% to 0.4% will not be detected, but may in fact be very significant for our business.
 

docs/source/user-guide/vision/tutorials/plot_classification_tutorial.py

@@ -196,7 +196,8 @@ def deepchecks_collate_fn(batch) -> BatchOutputFormat:
 test_data = VisionData(batch_loader=test_loader, task_type='classification', label_map=LABEL_MAP)
 #%%
 # Making sure our data is in the correct format:
-# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+#
 # The VisionData object automatically validates your data format and will alert you if there is a problem.
 # However, you can also manually view your images and labels to make sure they are in the correct format by using
 # the ``head`` function to conveniently visualize your data:
@@ -207,7 +208,8 @@ def deepchecks_collate_fn(batch) -> BatchOutputFormat:
 # And observe the output:
 #
 # Running Deepchecks' suite on our data and model!
-# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+#
 # Now that we have defined the task class, we can validate the train and test data with deepchecks' train test validation
 # suite.
 # This can be done with this simple few lines of code:

docs/source/user-guide/vision/tutorials/plot_custom_task_tutorial.py

@@ -176,7 +176,7 @@ def load_or_download(cls, root: Path, train: bool) -> 'CocoInstanceSegmentationD
 # predictions (when available). In order to do that, those must be in a pre-defined format, according to the task type.
 #
 # In the following example we're using pytorch. To see how this can be done using tensorflow or a generic generator,
-# please refer to :doc:`creating VisionData guide </user-guide/vision/VisionData#creating-a-visiondata-object>`.
+# please refer to :doc:`creating VisionData guide </user-guide/vision/VisionData>`.
 #
 # For pytorch, we will use our DataLoader, but we'll create a new collate function for it, that transforms the batch to
 # the correct format. Then, we'll create a :class:`deepchecks.vision.vision_data.vision_data.VisionData` object,

docs/source/user-guide/vision/tutorials/plot_detection_tutorial.py

@@ -292,7 +292,8 @@ def deepchecks_collate_fn(batch) -> BatchOutputFormat:
 
 #%%
 # Making sure our data is in the correct format:
-# ~~~~~~~~~~~~~~~~~~~~~~
+# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+#
 # The VisionData object automatically validates your data format and will alert you if there is a problem.
 # However, you can also manually view your images and labels to make sure they are in the correct format by using
 # the ``head`` function to conveniently visualize your data:

docs/source/user-guide/vision/tutorials/plot_segmentation_tutorial.py

@@ -132,7 +132,8 @@ def deepchecks_collate_fn(batch) -> BatchOutputFormat:
 
 #%%
 # Making sure our data is in the correct format:
-# ~~~~~~~~~~~~~~~~~~~~~~
+# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+#
 # The VisionData object automatically validates your data format and will alert you if there is a problem.
 # However, you can also manually view your images and labels to make sure they are in the correct format by using
 # the ``head`` function to conveniently visualize your data: