neka-nat / vin-keras

This is an implimentation of Value Iteration Networks (NIPS2016 best paper) in keras

vin-keras

This is an implimentation of VIN(Value Iteration Networks) in keras.

• 16x16 Grid world datasets from vin(chainer) repository

First, train and save VIN model.

$ python train.py

Next, predict path on the test data.

$ python predict.py

An example of the results are shown below.

Reference

• VIN(original)
• tensorflow-value-iteration-networks
• vin(chainer)

Requirements

• Keras
• python-opencv
• numpy
• joblib