The code is written in Python 3.6 in Ubuntu 18.04 Operating System.

********************************************************************************

Libraries required:

NumPy

PyTorch

********************************************************************************

Train the model by writing following in the terminal:

"python3 train.py opt1 opt2"

---------------------------------------

opt1: 1). '1' for bAbI Task training

      2). '2' for Medical NER task training

opt1: 1). 'GPU' to run code on GPU

      2). 'CPU' to run code normally

---------------------------------------

********************************************************************************

Test the model by writing following in the terminal:

"python3 test.py opt1 opt2 opt3 opt4"

---------------------------------------

opt1: 1). '1' for bAbI Task training

      2). '2' for Medical NER task training

opt2: 1). 'GPU' to run code on GPU

      2). 'CPU' to run code normally

opt3: Last Epoch number till the model was trained

Opt4: Last Batch Number till the model was trained 

---------------------------------------

********************************************************************************

Reference:

1). https://github.com/loudinthecloud/pytorch-ntm

2). Paper: Robust and Scalable Differentiable Neural Computer for Question Answering