Skip to content

Code for converting natural language instructions into its corresponding action sequence.

License

Notifications You must be signed in to change notification settings

pikaliov/listen-attend-and-walk

 
 

Repository files navigation

listen-attend-and-walk

Converting natural language instructions into its corresponding action sequence.

Here is the link to the base paper that is being referred.

Language used : Python 2.7

Prerequiisites

  • Anaconda - Anaconda includes all the Python-related dependencies
  • Theano - Computational graphs are built on Theano
  • ArgParse - Command line parsing in Python

Instructions

Train model

python train_model.py 

Test model

python test_model.py -fp PATH

PATH stands for the path to the file model.pkl which lies within the tracks folder. This folder is created by the end of training.

The SAIL route instruction dataset has been downloaded from here.

The following code has been used for embedding sentences to vectors. It uses the skip-gram model.

This project is built using Hongyuan Mei's code as base.

About

Code for converting natural language instructions into its corresponding action sequence.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 95.3%
  • Jupyter Notebook 4.7%