Implementation of a Recursive Neural Tensor Network using only python and numpy. This implentation follows the proposed model by Socher et al. in the paper Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank

Dependencies:

• python v2.7.x
• numpy v1.x

The optimizer implented to train the neural network is an AdaGrad optimizer as described in the paper.

For usage, please check the ipython notebook.