claminos / Unibo-NLP-Assignments

NLP assignments of the Artificial Intelligence course of study at the University of Bologna

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

NLP Assignments

Master in Artificial Intelligence - Natural Language Processing

Team Components

Michael Magdy Nasr Zaki Ghaly - michael.magdynasr@studio.unibo.it

Simona Scala - simona.scala6@studio.unibo.it

Sara Vorabbi - sara.vorabbi@studio.unibo.it

Overview

Assignment 1

The objective of the 1st assignment is to implement Recurrent Neural Networks (RNNs) for sequence labelling (PoS tagging) on the Penn Treebank dataset.

Assignment 2

The objective of the 2nd assignment is to implement Question Answering (QA) with transformers on CoQA dataset.

Repository structure and contents

    .
    ├── A1_PoS_Tagging_w_RNN                                        # folder of the 1st assignment
    │   ├── PoS_Tagging_w_RNN.ipynb                                 # notebook of the 1st assignment
    │   └── report_A1.pdf                                           # report of the 1st assignment
    ├── A2_Seq2Seq_Abstractive_Question_Answering_(QA)_on_CoQA      # folder of the 2nd assignment
    │   ├── bert_tiny_42.ipynb                                      # notebook of the model BERT2BERT on seed 42
    │   ├── bert_tiny_1337.ipynb                                    # notebook of the model BERT2BERT on seed 1337
    │   ├── bert_tiny_2022.ipynb                                    # notebook of the model BERT2BERT on seed 2022
    │   ├── distilroberta_42.ipynb                                  # notebook of the model RoBERTa2RoBERTa on seed 42
    │   ├── distilroberta_1337.ipynb                                # notebook of the model RoBERTa2RoBERTa on seed 1337
    │   ├── distilroberta_2022.ipynb                                # notebook of the model RoBERTa2RoBERTa on seed 2022
    │   └── report_A2.pdf                                           # report of the 2nd assignment
    ├── LICENSE
    └── README.md

About

NLP assignments of the Artificial Intelligence course of study at the University of Bologna


Languages

Language:Jupyter Notebook 100.0%