BDe score

[sonujose123]

I created BN from a BIF file (asia.bif) . But when I try to find the score using BDe with 'lizards.csv' it is failing.

Code -
from pyBN import * import numpy as np import os from os.path import dirname

file = 'data/asia.bif' bn = read_bn(file) dpath = os.path.join(dirname(dirname(dirname(dirname(file)))),'data')
path = (os.path.join(dpath,'lizards.csv')) data = np.loadtxt(path, dtype='int32',skiprows=1,delimiter=',')

print BDe(bn,data)

Error

Traceback (most recent call last): File "test.py", line 12, in print BDe(bn,data) File "/home/sonu/Documents/pyBN/pyBN-master/pyBN/learning/structure/score/bayes_scores.py", line 61, in BDe counts_dict = mle_fast(bn, data, counts=True, np=True) File "/home/sonu/Documents/pyBN/pyBN-master/pyBN/learning/parameter/mle.py", line 41, in mle_fast F[n]['values'] = list(nmp.unique(data[:,i])) IndexError: index 3 is out of bounds for axis 1 with size 3

[ncullen93]

I think this is because the network and the data don't match.. The andes network has these nodes: ['asia', 'smoke', 'tub', 'bronc', 'lung', 'either', 'dysp', 'xray'], whereas the lizards dataset only has 3 columns (nodes).. The BDe score measures how well a given BN fits a COMPATIBLE dataset (i.e. the nodes of the BN match up with the columns of the dataset). :)

[ncullen93]

Note, you can try LEARNING the structure from the lizards.csv dataset and check the BDe score.. OR you can generate your own random andes dataset and check the BDe score.

[sonujose123]

Thanks a lot for your quick response. On Dec 4, 2016 8:43 PM, "Nicholas Cullen" notifications@github.com wrote:

Note, you can try LEARNING the structure from the lizards.csv dataset and check the BDe score.. OR you can generate your own random andes dataset and check the BDe score.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ncullen93/pyBN/issues/20#issuecomment-264767026, or mute the thread https://github.com/notifications/unsubscribe-auth/AQjW7cACYYtAtT-aHoMcVKIBMBtBEL6yks5rE5aPgaJpZM4LD0wr .

[sonujose123]

Hi Nicholas, I tried with modified csv with asia.bif . Then also I am facing the same issue. I got this error - python test.py ['asia', 'smoke', 'tub', 'bronc', 'lung', 'either', 'dysp', 'xray'] ('asia', 'tub') ('smoke', 'bronc') ('smoke', 'lung') ('tub', 'either') ('bronc', 'dysp') ('lung', 'either') ('either', 'dysp') ('either', 'xray') /home/sonu/Documents/pyBN/pyBN-master/pyBN/learning/parameter/mle.py:48: VisibleDeprecationWarning: using a non-integer number instead of an integer will result in an error in the future F[rv]['cpt'] = nmp.histogram(data[:,rv], bins=bn.card(rv))[0] Traceback (most recent call last): File "test.py", line 20, in print BDe(bn,data) File "/home/sonu/Documents/pyBN/pyBN-master/pyBN/learning/structure/score/bayes_scores.py", line 61, in BDe counts_dict = mle_fast(bn, data, counts=True, np=True) File "/home/sonu/Documents/pyBN/pyBN-master/pyBN/learning/parameter/mle.py", line 48, in mle_fast F[rv]['cpt'] = nmp.histogram(data[:,rv], bins=bn.card(rv))[0] IndexError: only integers, slices (:), ellipsis (...), numpy.newaxis (None) and integer or boolean arrays are valid indices

Please help me.

Regards asia.bif.zip

Sonu lizards.csv.zip

[ncullen93]

Ok i think i fixed it... pandas must have changed their indexing since I wrote this. It should work if 'data' is a pandas dataframe whose columns are same as BN nodes.. but i think it will now be broken if data is numpy array.

[ncullen93]

Pull the repository and try again