def __init__(self,
split_type='cv',
partitions=10,
test_size=0.3,
seed=0,
fields=None):
if fields is None:
fields = ['X', 'Y']
config = self._to_config(locals())
# config cleaning.
if split_type == "cv":
del config['test_size']
elif split_type == "loo":
del config['partitions']
del config['partition']
del config['test']
del config['seed']
elif split_type == 'holdout':
pass
else:
raise Exception('Wrong split_type: ', split_type)
super().__init__(config)
from pjml.macro import split
self.transformer = Chain(
Expand(), split(split_type, partitions, test_size, seed, fields))
def __mul__(self, other):
from pjml.tool.chain import Chain
from pjml.config.description.cs.chaincs import ChainCS
if isinstance(other, (Chain, ChainCS)):
return Chain(self, *other.components)
if isinstance(self, (Chain, ChainCS)):
return Chain(*self.components, other)
return Chain(self, other)
def __init__(self,
split_type: str = "cv",
partitions: int = 10,
test_size: float = 0.3,
seed: int = 0,
fields: str = "X,Y",
**kwargs):
config = self._to_config(locals())
# config cleaning.
if split_type == "cv":
del config["test_size"]
elif split_type == "loo":
del config["partitions"]
del config["partition"]
del config["test"]
del config["seed"]
elif split_type == "holdout":
pass
else:
raise Exception("Wrong split_type: ", split_type)
from pjml.macro import split
self._component = Chain(
Repeat(), split(split_type, partitions, test_size, seed, fields))
super().__init__(config, **kwargs)
def printing_test(arq="iris.arff"):
print(Chain(Map(select(File(arq)))))
exp = Workflow(
File(arq),
Partition(),
Map(PCA(), SVMC(), Metric(enhance=False)),
Map(Report("<---------------------- fold"), enhance=False),
Summ(function="mean", enhance=False),
Reduce(),
Report("mean ... S: $S", enhance=False),
)
print(exp)
print(select(DT(), SVMC()))
sel = select(DT(), SVMC())
print(sel)
print(Map(DT()))
exp = ChainCS(
File(arq),
Partition(),
Map(PCA(), select(SVMC(), DT(criterion="gini")),
Metric(enhance=False)),
Report("teste"),
Map(Report("<---------------------- fold")),
)
print(exp)
def __init__(
self,
split_type: str = "holdout",
partitions: int = 2,
partition: int = 0,
test_size: float = 0.3,
seed: int = 0,
fields: str = "X,Y",
**kwargs,
):
config = self._to_config(locals())
trsplit = TrSplit(
split_type=split_type,
partitions=partitions,
partition=partition,
test_size=test_size,
seed=seed,
fields=fields,
**kwargs,
)
tssplit = TsSplit(
split_type=split_type,
partitions=partitions,
partition=partition,
test_size=test_size,
seed=seed,
fields=fields,
**kwargs,
)
# HINT: Chain should be in the order below; otherwise, input data will differ for trsplit and tssplit.
self._component = Chain(tssplit, trsplit)
super().__init__(config, **kwargs)
def __init__(self, config, seed, transformers, deterministic):
super().__init__(config, seed, transformers, deterministic)
# Implementation-wise, Container1(Chain(a,b,c)) is needed to make
# Container1(a,b,c) possible.
if len(self.transformers) > 1:
from pjml.tool.chain import Chain
self.transformer = Chain(transformers=self.transformers)
else:
self.transformer = self.transformers[0]
class Partition(HeavyTransformer):
"""Class to perform, e.g. Expand+kfoldCV.
This task is already done by function split(),
but if performance becomes a concern, this less modular solution is a
good choice.
TODO: the current implementation is just an alias for the nonoptimized
previous solution.
"""
def __init__(self,
split_type='cv',
partitions=10,
test_size=0.3,
seed=0,
fields=None):
if fields is None:
fields = ['X', 'Y']
config = self._to_config(locals())
# config cleaning.
if split_type == "cv":
del config['test_size']
elif split_type == "loo":
del config['partitions']
del config['partition']
del config['test']
del config['seed']
elif split_type == 'holdout':
pass
else:
raise Exception('Wrong split_type: ', split_type)
super().__init__(config)
from pjml.macro import split
self.transformer = Chain(
Expand(), split(split_type, partitions, test_size, seed, fields))
def _apply_impl(self, data):
splitter_model = self.transformer.apply(data)
applied = splitter_model.data.last_transformations_replaced(
drop=self.transformer.size,
transformation=self.transformations('a')[0])
return Model(self, data, applied, splitter_model=splitter_model)
def _use_impl(self, data, splitter_model=None):
used = splitter_model.use(data)
return used.last_transformations_replaced(
drop=self.transformer.size,
transformation=self.transformations('u')[0])
@classmethod
def _cs_impl(cls):
raise NotImplementedError
def __init__(self, config, enhancer_cls, model_cls, seed, components,
enhance, model, deterministic):
super().__init__(config, enhancer_cls, model_cls, seed, components,
enhance, model, deterministic)
# Implementation-wise, Container1(Chain(a,b,c)) is needed to make
# Container1(a,b,c) possible.
if len(self.components) > 1:
from pjml.tool.chain import Chain
self.component = Chain(components=self.components)
else:
self.component = self.components[0]
OnlyUse(Calc(functions=['flatten'])),
OnlyUse(Report('flatten S ... S: $S')),
OnlyUse(Calc(functions=['mean'])),
OnlyUse(Report('mean S ... S: $S')),
Report('End ...\n'),
)
# diversidade,
# Lambda(function='$R[0][0] * $R[0][1]', field='r')
print('sample .................')
pipe = full(rnd(expr, n=5), field='S', n=1).sample()
#
# pipes = rnd(expr, n=5)
#
# magia = Multi(pipes) -> Diversity() -> Agrega()
# magia.apply()
# coll = magia.use()
#
# pipe = full(pipes, field='S', n=1).sample()
print('apply .................')
data = Workflow(File("abalone3.arff"), Binarize()).apply().data
c = Chain(pipe.wrapped, Report())
model = c.apply(data)
print('use .................')
dataout = model.use(data)
def sample(self):
import numpy as np
from pjml.tool.chain import Chain
css = self.components.copy()
np.random.shuffle(css)
return Chain(transformers=[cs.sample() for cs in css])
def sample(self):
components = [cs.sample() for cs in self.components]
from pjml.tool.chain import Chain
return Chain(components=components)
),
Metric(functions=['accuracy'])
)
)
)
# {history.last.config['function']}
print(expr)
print('sample .................')
pipe = full(rnd(expr, n=10), field='S').sample()
pipe.enable_pretty_printing()
print(f'Pipe:\n{pipe}')
print(f'Wrapped:\n{pipe.unwrap}')
pipe = Chain(File('abalone3.arff'), Binarize(), Split(), pipe.unwrap,
Metric(), Report())
print('apply .................')
model = pipe.apply()
# print(222222222222222, dataout.history)
# data morre no apply() do predictor
print('use .................')
# print(3333333333333333, dataout.history)
# RUS desaparece no use()
exit(0)
def sample(self):
transformers = [cs.sample() for cs in self.components]
from pjml.tool.chain import Chain
return Chain(transformers=transformers)
def evaluator(*components, function="mean_std", **validation_args):
return Chain(Partition(**validation_args), Map(components=components),
Summ(function=function))
def evaluator(*components, function='mean_std', **validation_args):
return Chain(Partition(**validation_args), Map(transformers=components),
Summ(function=function))