Skip to content

Latest commit

 

History

History
29 lines (23 loc) · 1.26 KB

README.md

File metadata and controls

29 lines (23 loc) · 1.26 KB

thesis_CSAI

This repository contains the code Jupyter Notebook of the thesis: Evaluating Sentence Embeddings on Syntactic Information Through Representational Similarity Analysis.

The notebook expects the Penn Treebank data set, however, this data set is not publicly available.

The notebook is dependent on the following packages:

Further, the analysis in the notebook is dependent on the pre-trained BERT model and the pre-trained uni-skip and bi-skip models.

The BERT model can be downloaded via the following link: https://storage.googleapis.com/bert_models/2018_10_18/uncased_L-12_H-768_A-12.zip Store the contents in a folder named: "uncased_L-12_H-768_A-12"

The Skip-Thought models can be download via the following instructions: https://github.com/tensorflow/models/tree/master/research/skip_thoughts#download-pretrained-models-optional Store the contents in a folder named: "skipthought"