Task Description

• describe the input and output type and the required shape of the predictor
• describe if predictor returns 2D tensor -> classification scenario else if
  returns 1D tensor -> regression scenario
• correct the mistakes in the example code that not using the to_feat_stats():

        >>> training_data = ms.Tensor(np.random.rand(10, 4), ms.float32)
        >>> # initialize LIMETabular explainer with the model and training data
        >>> lime = LIMETabular(net, training_data, feature_names=feature_names, class_names=class_names)

• update the docstrings for issue #I4UXQS:[finic] Modify the LIMETabular interface

Task Goal

Sub Task