CFQIRBM provides Restricted Boltzmann Machine implementations to work on D-Wave quantum annealers. qrbm_reconstruct_img.ipynb shows how to use the RBM to reconstruct images. Further improvements we are considering to add to the project include

• Collaborative filtering usage
• Infinite Restricted Boltzmann Machine feature

Packages required are in requirements.txt

pip install -r requirements.txt

The following package should be downloaded from D-Wave cloud for contracts.

• dwave-sapi2==3.0

Demo in qrbm_reconstruct_img.ipynb

FIll in the following information in qrbm/sampler.py line 16 ~ 18 according to your D-Wave contracts.

self.endpoint = 'YOUR URL'
self.token = 'YOUR TOKEN'
self.solver = 'YOUR SOLVER'

• @shitian-ni

Please read the CLA below carefully before submitting your contribution.

https://www.mercari.com/cla/

Copyright 2018 Mercari, Inc.

Licensed under the Apache License 2.0.

https://en.wikipedia.org/wiki/Restricted_Boltzmann_machine