Exploring fastai’s Tabular Learner
Deep learning is not often associated with tabular or structured data, as this kind of data comes with some possible questions:
- How should we represent features in a way that can be understood by the neural networks? In tabular data, we often deal with numerical and categorical features, so we need to correctly represent both types of inputs.
- How do we use feature interactions, both between the features themselves and the target?
- How do we effectively sample the data? Tabular datasets tend to be smaller than typical datasets used for solving computer vision or NLP problems. There is no easy way to apply augmentation, such as random cropping or rotation in the case of images. Also, there is no general large dataset with some universal properties, based on which we could easily apply transfer learning.
- How do we interpret the predictions of a neural network?
That is why practitioners tend to use traditional...
