other pyAgrum.lib modules

bn2roc

pyAgrum.lib.bn2roc.module_help(exit_value=1, message='')

defines help viewed if args are not OK on command line, and exit with exit_value

pyAgrum.lib.bn2roc.showROC(bn, csv_name, variable, label, visible=True, show_fig=False, with_labels=True)

Compute the ROC curve and save the result in the folder of the csv file.

Parameters:
  • bn (pyAgrum.BayesNet) – a bayesian network
  • csv_name (str) – a csv filename
  • target (str) – the target
  • label (str) – the target label
  • visible (bool) – indicates if the resulting curve must be printed

bn2csv

Samples generation w.r.t to a probability distribution represented by a Bayesian network.

class pyAgrum.lib.bn2csv.CSVGenerator

Bases: object

Class for samples generation w.r.t to a probability distribution represented by a Bayesian network.

caching_probas(bn, node_id, n, par)
Parameters:
  • bn (pyAgrum.BayesNet) – a Bayesian network
  • node_id (int) – a node id
  • n (int) – a node id
  • par (list) – the node’s parents
Returns:

the node’s probabilities

Return type:

list

cachingnameAndParents(bn, n)

Compute a list of parents for node n in BN bn.

Parameters:
  • bn (pyAgrum.BayesNet) – a Bayesian network
  • n – (int) a node id
  • n – (str) a node name
Returns:

a couple of name of n and list of parents names

Return type:

tuple

static draw(tab)

draw a value using tab as probability table.

Parameters:tab (list) – a probability table
Returns:the couple (i,proba)
Return type:tuple
static nameAndParents(bn, n)

Compute a list of parents for node n in BN bn.

Parameters:
  • bn (pyAgrum.BayesNet) – a Bayesian network
  • n – (int) a node id
  • n – (str) a node name
Returns:

a couple of name of n and list of parents names

Return type:

tuple

Raises:

gum.IndexError – If the node is not in the Bayesian network

newSample(bn, seq)

Generate a sample w.r.t to the bn using the variable sequence seq (topological order)

Parameters:
  • bn (pyAgrum.BayesNet) – a Bayesian network
  • seq (list) – a variable sequence
Returns:

the coule (sample,log2-likelihood)

Return type:

tuple

proceed(name_in, name_out, n, visible, with_labels)

From the file name_in (BN file), generate n samples and save them in name_out

Parameters:
  • name_in (str) – a file name
  • name_out (str) – the output file
  • n (int) – the number of samples
  • visible (bool) – indicate if a progress bar should be displayed
  • with_labels (bool) – indicate if values should be labelled or not
Returns:

the log2-likelihood of the n samples database

Return type:

double

pyAgrum.lib.bn2csv.generateCSV(name_in, name_out, n, visible=False, with_labels=True)

From the file name_in (BN file), generate n samples and save them in name_out

Parameters:
  • name_in (str) – a file name
  • name_out (str) – the output file
  • n (int) – the number of samples
  • visible (bool) – indicate if a progress bar should be displayed
  • with_labels (bool) – indicate if values should be labelled or not
Returns:

the log2-likelihood of the n samples database

Return type:

double

pyAgrum.lib.bn2csv.module_help(exit_value=1)

defines help viewed if args are not OK on command line, and exit with exit_value

bn2scores

pyAgrum.lib.bn2scores.checkCompatibility(bn, fields, csv_name)

check if variables of the bn are in the fields

if not : return None if compatibilty : return a list of position for variables in fields

pyAgrum.lib.bn2scores.computeScores(bn_name, csv_name, visible=False, transforme_label=None)
pyAgrum.lib.bn2scores.getNumLabel(inst, i, label, transforme_label)
pyAgrum.lib.bn2scores.lines_count(filename)

count lines in a file

pyAgrum.lib.bn2scores.module_help(exit_value=1)

defines help viewed if args are not OK on command line, and exit with exit_value

pyAgrum.lib.bn2scores.stringify(s)

bn_vs_bn

pyAgrum.lib.bn_vs_bn.compareBN(name1, name2)
Parameters:
  • name1 (str) – a BN filename
  • name2 (str) – another BN filename
Returns:

“OK” if bn are the same, a description of the error otherwise

Return type:

str

pyAgrum.lib.bn_vs_bn.compareBNCPT(b1, b2)
Parameters:
  • b1 (pyAgrum.BayesNet) – a Bayesian network
  • b2 (pyAgrum.BayesNet) – another Bayesian network
Returns:

‘OK’ if b2 have (at least) the same variable as b1 and their cpts are the same

Return type:

str

pyAgrum.lib.bn_vs_bn.compareBNParents(b1, b2)
Parameters:
  • b1 (pyAgrum.BayesNet) – a Bayesian network
  • b2 (pyAgrum.BayesNet) – another Bayesian network
Returns:

‘OK’ if b2 have (at least) the same variable as b1 and their parents are the same.

Return type:

str

pyAgrum.lib.bn_vs_bn.compareBNVariables(b1, b2)
Parameters:
  • bn1 (pyAgrum.BayesNet) – a Bayesian network
  • bn2 (pyAgrum.BayesNet) – another Bayesian network
Returns:

‘OK’ if BN are composed of the same variables, indicates a non-existing variable otherwise

Return type:

str

pyAgrum.lib.bn_vs_bn.compareCPT(b1, cpt1, b2, cpt2)
Parameters:
  • b1 (pyAgrum.BayesNet) – a Bayesian network
  • cpt1 (pyAgrum.Potential) – one of b1’s cpts
  • b2 (pyAgrum.BayesNet) – another Bayesian network
  • cpt2 (pyAgrum.Potential) – one of b2’s cpts
Returns:

‘OK’ if CPTs are the same

Return type:

str

Raises:

gum.KeyError – If cpts are not from the same variable

pyAgrum.lib.bn_vs_bn.module_help(exit_value=1)

defines help viewed if args are not OK on command line, and exit with exit_value

pyAgrum.lib.bn_vs_bn.nodeId(bn, n)
Parameters:
  • bn (pyAgrum.BayesNet) – a Bayesian network
  • n (str) – the name of the node
Returns:

the id of the node

Return type:

int

Raises:

gum.IndexError – If the node is not in the Bayesian network

pyAgrum.lib.bn_vs_bn.parents_name(bn, n)
Parameters:
  • bn (pyAgrum.BayesNet) – a Bayesian network
  • n – (str) the name of the node
  • n – (int) the id of the node
Returns:

a list of name of parents of node n

Return type:

map

Raises:

gum.IndexError – If the node is not in the Bayesian network

pretty_print

pyAgrum.lib.pretty_print.bn2txt(aBN)

Representation of all CPTs of a gum.BayesNet

Parameters:aBN – the bayes net or the name of the file
Returns:
pyAgrum.lib.pretty_print.cpt2txt(cpt, digits=4)

string representation of a gum.Potential

Parameters:cpt – the Potential to represent
Returns:the string representation
pyAgrum.lib.pretty_print.max_length(v)
pyAgrum.lib.pretty_print.module_help(exit_value=1)

defines help viewed if args are not OK on command line, and exit with exit_value