artiste-qb-net / quantum-fog Goto Github PK

https://qbnets.wordpress.com/2016/03/22/first-version-of-qubiter-a-quantum-computer-simulator-is-out-it-says-welcome-my-navigator-where-do-you-want-me-to-take-you-next/

https://qbnets.wordpress.com/2016/08/14/quantum-fog-close-to-becoming-sentient-it-can-now-distinguish-between-the-words-good-and-evil/

https://qbnets.wordpress.com/2017/03/20/bnt-and-pnl-two-masterpieces-of-bayesian-networks-retro-art/

https://qbnets.wordpress.com/2017/02/11/r-are-us-we-are-all-r-now/

https://qbnets.wordpress.com/2016/09/20/in-love-with-jupyter-notebooks-post-processing-your-lab-notebook/

In my branch I extended the EnumerationEngine class to accept an additional argument print_format (-> screenshot). It defaults to text but when specifying HTML it produces a joint distribution table as shown in the attached screen-shot.

If you agree that this is a nice feature to have, I will incorporate this as I go along for all other print output that could be wrapped into a HTML table.

https://qbnets.wordpress.com/2017/02/22/quantum-fogs-weight-in-bnlearn-units/

On Linux, in pycharm, the code execution of the BayesNet.py example stops at the line bnet.draw(algo_num=2) and opens a viewer, thereby masking errors with the reading and writing sample files that follows after that. Nevertheless the code will get executed after closing the window. Not the behaviour I would have expected.

https://qbnets.wordpress.com/2016/02/17/our-baby-quantum-fog-can-now-read-write-and-draw/

https://qbnets.wordpress.com/2016/02/02/first-major-commit-to-quantum-fog-in-python/

https://qbnets.wordpress.com/2015/11/30/quantum-fog-comes-out-of-the-closet/

https://qbnets.wordpress.com/2018/02/09/today-enrolled-our-baby-quantum-fog-in-gambling-school-taught-by-famous-monte-carlo-gamblers-pymc3-edward-zhusuan/

https://qbnets.wordpress.com/2017/12/16/in-love-with-jupyter-widgets/

Drawing should default to graphviz module when available otherwise fall back on networkx.

Pulled the latest repo to retrace these steps. The networkx rendering works, but I get the following error message:

Traceback (most recent call last):
File "/home/quax/projects/quantum-fog/graphs/BayesNet.py", line 158, in
bnet.write_dot(path1)
File "/home/quax/projects/quantum-fog/graphs/Graph.py", line 258, in write_dot
nx.write_dot(self.get_nx_graph(), path)
AttributeError: module 'networkx' has no attribute 'write_dot'

Hi, I read your program about inference with edward, but I don't understand how to load a bayes network which is in a bif format.
Can you help me ?
THANK YOU!

https://qbnets.wordpress.com/2016/07/14/quantum-fog-a-quantum-computer-simulator-based-on-quantum-bayesian-networks-can-now-think-at-least-better-than-a-rock/

I created a simple Monty Hall B-net BIF. It contains an error in the potential definition for the Monty_Opens node. This error is not caught when the BIF is loaded.

When running the Brute Force Enumeration Engine this creates unexpected results but no errors, fortunately the Join Tree Engine throws errors that show where this is coming from.

QFog should not accept an incorrect BIF file as valid input. This also merits some further investigation to see if one can create such an inconsistent graph through the API.

The corrupt/corrupting BIF file is attached.
Monty_Hall.txt

https://qbnets.wordpress.com/2016/02/26/quantum-fog-gets-prefab-nodes-prefab-but-of-high-quality-i-assure-you/