Expanding abstract method documentation

As per #483 (comment)

Author: dumbledad

InnerEye/ML/deep_learning_config.py

@@ -303,17 +303,23 @@ def get_cross_validation_hyperdrive_sampler(self) -> GridParameterSampling:
     @abstractmethod
     def get_cross_validation_hyperdrive_config(self, run_config: ScriptRunConfig) -> HyperDriveConfig:
         """
-        Is implemented in both ModelConfigBase and LightningContainer
-        :param run_config: AzureML estimator
-        :return: HyperDriveConfigs
+        Returns a configuration for AzureML Hyperdrive that varies the cross validation split index.
+        
+        This abstract method is implemented in both ModelConfigBase and LightningContainer
+        :param run_config: The AzureML run configuration object that configures training for an individual model.
+        :return: A hyperdrive configuration object.
         """
         pass
 
     @abstractmethod
     def get_parameter_search_hyperdrive_config(self, run_config: ScriptRunConfig) -> HyperDriveConfig:
         """
-        Is implemented in ModelConfigBase but not in LightningContainer. Classes inheriting
-        from LightningCountainer may wish to provide an implementation.
+        Turns the given AzureML estimator (settings for running a job in AzureML) into a configuration object
+        for doing hyperparameter searches.
+
+        This abstract method is implemented in ModelConfigBase but not in LightningContainer. Classes inheriting
+        from LightningCountainer will wish to provide an implementation if they provide hyper drive support for
+        parameter search.
         :param run_config: AzureML estimator
         :return: HyperDriveConfigs
         """