The objective of this assignment is to build both CNN and RNN model for predicting relationship given two pieces of text, a premise and a hypothesis: the premise entails, contradicts or is neutral to the hypothesis.
The SNLI dataset includes 100000 examples in the training set and 1000 examples in the validation set. In this assignment, we will build a CNN model with 2-layer 1-D convolutional layer with ReLU activations and a RNN model with a single-layer, bi-directional GRU, perform hyperparameters-tuning on SNLI dataset and evaluate on MultiNLI dataset.