def mine_c45(table, result):
""" An entry point for C45 algorithm.
_table_ - a dict representing data table in the following format:
{
"<column name>': [<column values>],
"<column name>': [<column values>],
...
}
_result_: a string representing a name of column indicating a result.
"""
col = max([(k, gain(table, k, result))
for k in table.keys() if k != result],
key=lambda x: x[1])[0]
tree = []
for subt in get_subtables(table, col):
v = subt[col][0]
if is_mono(subt[result]):
tree.append(
['%s=%s' % (col, v),
'%s=%s' % (result, subt[result][0])])
else:
del subt[col]
tree.append(['%s=%s' % (col, v)] + mine_c45(subt, result))
return tree
def infox(table, col, res_col):
""" Calculates the entropy of the table _table_
after dividing it on the subtables by column _col_.
"""
s = 0 # sum
for subt in utils.get_subtables(table, col):
s += (float(len(subt[col])) / len(table[col])) * info(subt, res_col)
return s
def infox(table, col, res_col):
""" Calculates the entropy of the table _table_
after dividing it on the subtables by column _col_.
"""
s = 0 # sum
for subt in utils.get_subtables(table, col):
s += (float(len(subt[col])) / len(table[col])) * info(subt, res_col)
return s
def test_get_subtables(self):
expected = [{
'result': ['yes', 'no'],
'arg1': ['left', 'left'],
'arg2': ['down', 'up'],
'arg3': ['no', 'yes'],
}, {
'result': ['yes', 'no'],
'arg1': ['right', 'right'],
'arg2': ['down', 'down'],
'arg3': ['yes', 'no'],
}]
self.assertEqual(utils.get_subtables(self.table, 'arg1'), expected)
def test_get_subtables(self):
expected = [
{ 'result': ['yes', 'no'],
'arg1': ['left', 'left'],
'arg2': ['down', 'up'],
'arg3': ['no', 'yes'],
},
{ 'result': ['yes', 'no'],
'arg1': ['right', 'right'],
'arg2': ['down', 'down'],
'arg3': ['yes', 'no'],
}]
self.assertEqual(utils.get_subtables(self.table, 'arg1'), expected)
def mine_c45(table, result):
col = max([(k, gain(table, k, result)) for k in table.keys() if k != result],
key=lambda x: x[1])[0]
tree = []
for subt in get_subtables(table, col):
v = subt[col][0]
if is_mono(subt[result]):
tree.append(['%s=%s' % (col, v),
'%s=%s' % (result, subt[result][0])])
else:
del subt[col]
tree.append(['%s=%s' % (col, v)] + mine_c45(subt, result))
return tree
def mine_c45(table, result):
col = max([(k, gain(table, k, result))
for k in table.keys() if k != result],
key=lambda x: x[1])[0]
tree = []
for subt in get_subtables(table, col):
v = subt[col][0]
if is_mono(subt[result]):
tree.append(
['%s=%s' % (col, v),
'%s=%s' % (result, subt[result][0])])
else:
del subt[col]
tree.append(['%s=%s' % (col, v)] + mine_c45(subt, result))
return tree
def mine_c45(table, result):
""" An entry point for C45 algorithm.
_table_ - a dict representing data table in the following format:
{
"<column name>': [<column values>],
"<column name>': [<column values>],
...
}
_result_: a string representing a name of column indicating a result.
"""
tree = []
# Special case when there is a mixed strategy
if len(table.keys()) == 1:
key_distr = get_distribution(table[result])
for k in key_distr[1].keys():
tree.append([
'probability=%f' % (key_distr[1][k] / key_distr[0]),
'%s=%s' % (result, k)
])
return tree
# All other cases
col = max([(k, gain(table, k, result))
for k in table.keys() if k != result],
key=lambda x: x[1])[0]
for subt in get_subtables(table, col):
v = subt[col][0]
if is_mono(subt[result]):
tree.append(
['%s=%s' % (col, v),
'%s=%s' % (result, subt[result][0])])
else:
del subt[col]
tree.append(['%s=%s' % (col, v)] + mine_c45(subt, result))
return tree
def mine_c45(table, result):
""" An entry point for C45 algorithm.
_table_ - a dict representing data table in the following format:
{
"<column name>': [<column values>],
"<column name>': [<column values>],
...
}
_result_: a string representing a name of column indicating a result.
"""
col = max([(k, gain(table, k, result)) for k in table.keys() if k != result],
key=lambda x: x[1])[0]
tree = []
for subt in get_subtables(table, col):
v = subt[col][0]
if is_mono(subt[result]):
tree.append(['%s=%s' % (col, v),
'%s=%s' % (result, subt[result][0])])
else:
del subt[col]
tree.append(['%s=%s' % (col, v)] + mine_c45(subt, result))
return tree
def mine_c45(table, result):
""" An entry point for C45 algorithm.
_table_ - a dict representing data table in the following format:
{
"<column name>': [<column values>],
"<column name>': [<column values>],
...
}
_result_: a string representing a name of column indicating a result.
"""
tree = []
# Special case when there is a mixed strategy
if len(table.keys()) == 1:
key_distr = get_distribution(table[result])
for k in key_distr[1].keys():
tree.append(['probability=%f' % (key_distr[1][k] / key_distr[0]),
'%s=%s' % (result, k)])
return tree
# All other cases
col = max([(k, gain(table, k, result)) for k in table.keys() if k != result],
key=lambda x: x[1])[0]
for subt in get_subtables(table, col):
v = subt[col][0]
if is_mono(subt[result]):
tree.append(['%s=%s' % (col, v),
'%s=%s' % (result, subt[result][0])])
else:
del subt[col]
tree.append(['%s=%s' % (col, v)] + mine_c45(subt, result))
return tree
def info_x(table, col, res_col):
return sum(
len(subtable[col]) / len(table[col]) * info(subtable, res_col)
for subtable in utils.get_subtables(table, col))
def test_get_subtables(self):
expected = [
{"result": ["yes", "no"], "arg1": ["left", "left"], "arg2": ["down", "up"], "arg3": ["no", "yes"]},
{"result": ["yes", "no"], "arg1": ["right", "right"], "arg2": ["down", "down"], "arg3": ["yes", "no"]},
]
self.assertEquals(utils.get_subtables(self.table, "arg1"), expected)
