From f62991d7ce7c3b0b8dcd866e5ff4496c5ca6df74 Mon Sep 17 00:00:00 2001
From: Baptiste Bauvin <baptiste.bauvin@lis-lab.fr>
Date: Wed, 25 Jul 2018 10:22:00 -0400
Subject: [PATCH] Modified distrib of HPs for CQB and CQBv2
---
.../Monoview/ExecClassifMonoView.py | 2 +-
.../Monoview/MonoviewUtils.py | 24 +++++++++++++++++++
.../MonoviewClassifiers/CQBoost.py | 14 +++++++----
.../MonoviewClassifiers/CQBoostv2.py | 22 +++++++++--------
4 files changed, 47 insertions(+), 15 deletions(-)
diff --git a/multiview_platform/MonoMultiViewClassifiers/Monoview/ExecClassifMonoView.py b/multiview_platform/MonoMultiViewClassifiers/Monoview/ExecClassifMonoView.py
index 14398878..e0d532dc 100644
--- a/multiview_platform/MonoMultiViewClassifiers/Monoview/ExecClassifMonoView.py
+++ b/multiview_platform/MonoMultiViewClassifiers/Monoview/ExecClassifMonoView.py
@@ -161,7 +161,7 @@ def getHPs(classifierModule, hyperParamSearch, nIter, CL_type, X_train, y_train,
outputFileName, classifierModule,
KFolds=KFolds, nbCores=nbCores,
metric=metrics[0], nIter=nIter)
- logging.debug("Done:\t " + hyperParamSearch + "RandomSearch best settings")
+ logging.debug("Done:\t " + hyperParamSearch + " best settings")
else:
clKWARGS = kwargs[CL_type + "KWARGS"]
return clKWARGS, testFoldsPreds
diff --git a/multiview_platform/MonoMultiViewClassifiers/Monoview/MonoviewUtils.py b/multiview_platform/MonoMultiViewClassifiers/Monoview/MonoviewUtils.py
index dd7c33fa..30a342c3 100644
--- a/multiview_platform/MonoMultiViewClassifiers/Monoview/MonoviewUtils.py
+++ b/multiview_platform/MonoMultiViewClassifiers/Monoview/MonoviewUtils.py
@@ -1,5 +1,6 @@
from sklearn.model_selection import RandomizedSearchCV
import numpy as np
+from scipy.stats import uniform, randint
from .. import Metrics
from ..utils import HyperParameterSearch
@@ -50,6 +51,29 @@ def genTestFoldsPreds(X_train, y_train, KFolds, estimator):
testFoldsPreds = np.array([testFoldPreds[:minFoldLength] for testFoldPreds in testFoldsPreds])
return testFoldsPreds
+
+class CustomRandint:
+ def __init__(self, low=0, high=0, multiplier="e-"):
+ self.randint = randint(low, high)
+ self.multiplier = multiplier
+
+ def rvs(self, random_state=None):
+ randinteger = self.randint.rvs(random_state=random_state)
+ if self.multiplier == "e-":
+ return 10 ** -randinteger
+
+
+class CustomUniform:
+ def __init__(self, loc=0, state=1, multiplier="e-"):
+ self.uniform = uniform(loc, state)
+ self.multiplier = multiplier
+
+ def rvs(self, random_state=None):
+ unif = self.uniform.rvs(random_state=random_state)
+ if self.multiplier == 'e-':
+ return 10 ** -unif
+
+
# def isUseful(labelSupports, index, CLASS_LABELS, labelDict):
# if labelSupports[labelDict[CLASS_LABELS[index]]] != 0:
# labelSupports[labelDict[CLASS_LABELS[index]]] -= 1
diff --git a/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoost.py b/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoost.py
index b53ab517..0d1ecb15 100644
--- a/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoost.py
+++ b/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoost.py
@@ -19,9 +19,11 @@ from sklearn.base import BaseEstimator, ClassifierMixin
from sklearn.pipeline import Pipeline
from sklearn.model_selection import RandomizedSearchCV
from sklearn.tree import DecisionTreeClassifier
-from scipy.stats import randint
+from scipy.stats import randint, uniform
import numpy as np
+from ..Monoview.MonoviewUtils import CustomUniform, CustomRandint
+
class ColumnGenerationClassifier(BaseEstimator, ClassifierMixin):
@@ -293,6 +295,10 @@ class CQBoost(CqBoostClassifier):
return interpretString
+
+
+
+
def canProbas():
return False
@@ -311,7 +317,7 @@ def paramsToSet(nIter, randomState):
"""Used for weighted linear early fusion to generate random search sets"""
paramsSet = []
for _ in range(nIter):
- paramsSet.append({"mu": randomState.uniform(1e-02, 10**(-0.5)),
+ paramsSet.append({"mu": 10**-randomState.uniform(0.5, 1.5),
"epsilon": 10**-randomState.randint(1, 15),
"n_max_iterations": None})
return paramsSet
@@ -331,8 +337,8 @@ def genPipeline():
def genParamsDict(randomState):
- return {"classifier__mu": [0.001, 0.002],
- "classifier__epsilon": [1e-08, 2e-08],
+ return {"classifier__mu": CustomUniform(loc=.5, state=2, multiplier='e-'),
+ "classifier__epsilon": CustomRandint(low=1, high=15, multiplier='e-'),
"classifier__n_max_iterations": [None]}
diff --git a/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoostv2.py b/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoostv2.py
index 0039a4f4..839cd7c6 100644
--- a/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoostv2.py
+++ b/multiview_platform/MonoMultiViewClassifiers/MonoviewClassifiers/CQBoostv2.py
@@ -7,21 +7,23 @@ from collections import defaultdict, OrderedDict
import pandas as pd
import sys
from functools import partial
-import numpy as np
-from scipy.spatial import distance
from sklearn.base import BaseEstimator, ClassifierMixin, TransformerMixin
from sklearn.utils.validation import check_is_fitted
from sklearn.preprocessing import LabelEncoder
-from sklearn.tree import DecisionTreeClassifier
-from sklearn.metrics.pairwise import rbf_kernel, linear_kernel
-import numpy as np
from sklearn.base import BaseEstimator, ClassifierMixin
from sklearn.pipeline import Pipeline
-from sklearn.model_selection import RandomizedSearchCV
-from sklearn.tree import DecisionTreeClassifier
-from scipy.stats import randint
import numpy as np
+# import numpy as np
+# from scipy.spatial import distance
+# from sklearn.tree import DecisionTreeClassifier
+# from sklearn.metrics.pairwise import rbf_kernel, linear_kernel
+# from sklearn.model_selection import RandomizedSearchCV
+# from sklearn.tree import DecisionTreeClassifier
+# from scipy.stats import randint
+
+
+from ..Monoview.MonoviewUtils import CustomRandint, CustomUniform
class ColumnGenerationClassifierv2(BaseEstimator, ClassifierMixin):
def __init__(self, epsilon=1e-06, n_max_iterations=None, estimators_generator=None, dual_constraint_rhs=0, save_iteration_as_hyperparameter_each=None):
@@ -350,8 +352,8 @@ def genPipeline():
def genParamsDict(randomState):
- return {"classifier__mu": [0.001, 0.002],
- "classifier__epsilon": [1e-08, 2e-08],
+ return {"classifier__mu": CustomUniform(loc=.5, state=2, multiplier='e-'),
+ "classifier__epsilon": CustomRandint(low=1, high=15, multiplier='e-'),
"classifier__n_max_iterations": [None]}
--
GitLab