Skip to content

dtak/dynamic-mixing

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Kernelised Dynamic Mixing

Parbhoo, Gottesman et al 2018

A collection of methods for dynamically switching between kernel and model-based planning. Included in the 'toy' folder:

  • a script to create a toy data set of a certain size
  • training, test and validation data sets with corresponding fencepost sets indicating where each sequence starts and ends

To run KDM, first open the run.py file. In the get_parser() function, make sure that the default settings for the arguments '--dataset' and '--out_dir' are set to the location of the dataset and the directory to output to respectively.

Once the dataset and output directories have been set, run KDM by executing

python run.py

Alternatively, you can set these parameters by executing

python run.py --dataset d --out_dir o from the command line.

This will produce a list of optimal policies for the data set.

Please reach out to [email protected] for any bug reports or concerns. Original Paper at https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0205839