Zero Shot Learning is a way to be able to infer dataset's members without training on it. It is mostly achieved by some form of transfer learning, by which knowledge learned from one dataset can be applied on a different one. While people have proposed multiple zero shot learning approaches for vision tasks where knowledge from imagenet dataset can be used on new ones, we haven't yet seen any example of zero shot learning for text classification. In our latest research work, we have proposed a method to do zero shot learning on text, where an algorithm trained to learn relationships between sentences and their categories on a large noisy dataset can be made to generalize to new categories or even new datasets. We call the paradigm "Train Once , Test Anywhere". We also propose multiple neural network algorithms that can take advantage of this training methodology and get good results on different datasets. The best method uses an LSTM model for the task of learning relationships. The idea is if one can model the concept of "belongingness" between sentences and classes, the knowledge is useful for unseen classes or even unseen datasets.