From b28de9bc7beceaf145e9722a51ce8164dbc81926 Mon Sep 17 00:00:00 2001
From: Baptiste Bauvin <baptiste.bauvin@lis-lab.fr>
Date: Wed, 24 Mar 2021 11:08:21 -0400
Subject: [PATCH] Added multvew algorithms
---
config_files/config_test.yml | 4 +-
copyright.py | 106 ++++++++++++++++++
copyrightstamp.txt | 36 ++++++
license.txt | 30 +++++
.../config_files/config_example_0.yml | 4 +-
.../monoview_classifiers/sgd.py | 4 +-
.../additions/early_fusion_from_monoview.py | 68 +++++++++++
.../early_fusion_adaboost.py | 23 ++++
.../early_fusion_decision_tree.py | 22 ++++
.../early_fusion_gradient_boosting.py | 21 ++++
.../early_fusion_lasso.py | 17 +++
.../early_fusion_random_forest.py | 23 ++++
.../multiview_classifiers/early_fusion_sgd.py | 19 ++++
.../early_fusion_svm_rbf.py | 13 +++
14 files changed, 384 insertions(+), 6 deletions(-)
create mode 100644 copyright.py
create mode 100644 copyrightstamp.txt
create mode 100644 license.txt
create mode 100644 summit/multiview_platform/multiview_classifiers/additions/early_fusion_from_monoview.py
create mode 100644 summit/multiview_platform/multiview_classifiers/early_fusion_adaboost.py
create mode 100644 summit/multiview_platform/multiview_classifiers/early_fusion_decision_tree.py
create mode 100644 summit/multiview_platform/multiview_classifiers/early_fusion_gradient_boosting.py
create mode 100644 summit/multiview_platform/multiview_classifiers/early_fusion_lasso.py
create mode 100644 summit/multiview_platform/multiview_classifiers/early_fusion_random_forest.py
create mode 100644 summit/multiview_platform/multiview_classifiers/early_fusion_sgd.py
create mode 100644 summit/multiview_platform/multiview_classifiers/early_fusion_svm_rbf.py
diff --git a/config_files/config_test.yml b/config_files/config_test.yml
index 478e0e08..7130ff97 100644
--- a/config_files/config_test.yml
+++ b/config_files/config_test.yml
@@ -21,9 +21,9 @@ split: 0.8
nb_folds: 2
nb_class: 3
classes:
-type: [ "monoview"]
+type: ["multiview"]
algos_monoview: ["decision_tree", ]
-algos_multiview: ["weighted_linear_late_fusion"]
+algos_multiview: ["early_fusion_adaboost"]
stats_iter: 3
metrics:
accuracy_score: {}
diff --git a/copyright.py b/copyright.py
new file mode 100644
index 00000000..f471123b
--- /dev/null
+++ b/copyright.py
@@ -0,0 +1,106 @@
+# -*- coding: utf-8 -*-
+from __future__ import print_function, division
+import time
+import os
+import sys
+import fileinput
+
+
+def findFiles(directory, files=[]):
+ """scan a directory for py, pyx, pxd extension files."""
+ for filename in os.listdir(directory):
+ path = os.path.join(directory, filename)
+ if os.path.isfile(path) and (path.endswith(".py") or
+ path.endswith(".pyx") or
+ path.endswith(".pxd")):
+ if filename != "__init__.py" and filename != "version.py":
+ files.append(path)
+ elif os.path.isdir(path):
+ findFiles(path, files)
+ return files
+
+
+def fileUnStamping(filename):
+ """ Remove stamp from a file """
+ is_stamp = False
+ for line in fileinput.input(filename, inplace=1):
+ if line.find("# COPYRIGHT #") != -1:
+ is_stamp = not is_stamp
+ elif not is_stamp:
+ print(line, end="")
+
+
+def fileStamping(filename, stamp):
+ """ Write a stamp on a file
+
+ WARNING : The stamping must be done on an default utf8 machine !
+ """
+ old_stamp = False # If a copyright already exist over write it.
+ for line in fileinput.input(filename, inplace=1):
+ if line.find("# COPYRIGHT #") != -1:
+ old_stamp = not old_stamp
+ elif line.startswith("# -*- coding: utf-8 -*-"):
+ print(line, end="")
+ print(stamp)
+ elif not old_stamp:
+ print(line, end="")
+
+
+def getStamp(date, multimodal_version):
+ """ Return the corrected formated stamp """
+ stamp = open("copyrightstamp.txt").read()
+ stamp = stamp.replace("DATE", date)
+ stamp = stamp.replace("MULTIMODAL_VERSION", multimodal_version)
+ stamp = stamp.replace('\n', '\n# ')
+ stamp = "# " + stamp
+ stamp = stamp.replace("# \n", "#\n")
+ return stamp.strip()
+
+
+def getVersionsAndDate():
+ """ Return (date, multimodal_version..
+ ) """
+ v_text = open('VERSION').read().strip()
+ v_text_formted = '{"' + v_text.replace('\n', '","').replace(':', '":"')
+ v_text_formted += '"}'
+ v_dict = eval(v_text_formted)
+ return (time.strftime("%Y"), v_dict['multimodal'])
+
+
+def writeStamp():
+ """ Write a copyright stamp on all files """
+ stamp = getStamp(*getVersionsAndDate())
+ files = findFiles(os.path.join(os.path.dirname(os.path.abspath(__file__)),
+ "multimodal"))
+ for filename in files:
+ fileStamping(filename, stamp)
+ fileStamping("setup.py", stamp)
+
+
+def eraseStamp():
+ """ Erase a copyright stamp from all files """
+ files = findFiles(os.path.join(os.path.dirname(os.path.abspath(__file__)),
+ "multimodal"))
+ for filename in files:
+ fileUnStamping(filename)
+ fileUnStamping("setup.py")
+
+
+def usage(arg):
+ print("Usage :")
+ print("\tpython %s stamping" % arg)
+ print("\tpython %s unstamping" % arg)
+
+
+if __name__ == "__main__":
+ if len(sys.argv) == 1:
+ usage(sys.argv[0])
+ elif len(sys.argv) == 2:
+ if sys.argv[1].startswith("unstamping"):
+ eraseStamp()
+ elif sys.argv[1].startswith("stamping"):
+ writeStamp()
+ else:
+ usage(sys.argv[0])
+ else:
+ usage(sys.argv[0])
diff --git a/copyrightstamp.txt b/copyrightstamp.txt
new file mode 100644
index 00000000..872fce0e
--- /dev/null
+++ b/copyrightstamp.txt
@@ -0,0 +1,36 @@
+######### COPYRIGHT #########
+
+Copyright(c) DATE
+-----------------
+
+
+* Université d'Aix Marseille (AMU) -
+* Centre National de la Recherche Scientifique (CNRS) -
+* Université de Toulon (UTLN).
+* Copyright © 2019-2020 AMU, CNRS, UTLN
+
+Contributors:
+------------
+
+* Sokol Koço <sokol.koco_AT_lis-lab.fr>
+* Cécile Capponi <cecile.capponi_AT_univ-amu.fr>
+* Dominique Benielli <dominique.benielli_AT_univ-amu.fr>
+* Baptiste Bauvin <baptiste.bauvin_AT_univ-amu.fr>
+
+Description:
+-----------
+
+
+
+Version:
+-------
+
+* multiview_generator version = MULTIMODAL_VERSION
+
+Licence:
+-------
+
+License: New BSD License
+
+
+######### COPYRIGHT #########
diff --git a/license.txt b/license.txt
new file mode 100644
index 00000000..573f99a8
--- /dev/null
+++ b/license.txt
@@ -0,0 +1,30 @@
+New BSD License
+
+Copyright (c) 2020-15-01, The scikit-multimodallearn developers.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+ a. Redistributions of source code must retain the above copyright notice,
+ this list of conditions and the following disclaimer.
+ b. Redistributions in binary form must reproduce the above copyright
+ notice, this list of conditions and the following disclaimer in the
+ documentation and/or other materials provided with the distribution.
+ c. Neither the name of the IntertwiningWavelet developers nor the names of
+ its contributors may be used to endorse or promote products
+ derived from this software without specific prior written
+ permission.
+
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE FOR
+ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+DAMAGE.
\ No newline at end of file
diff --git a/summit/examples/config_files/config_example_0.yml b/summit/examples/config_files/config_example_0.yml
index 6e0e029f..753e5c07 100644
--- a/summit/examples/config_files/config_example_0.yml
+++ b/summit/examples/config_files/config_example_0.yml
@@ -42,9 +42,9 @@ classes:
# The type of algorithms to run during the benchmark (monoview and/or multiview)
type: ["monoview","multiview"]
# The name of the monoview algorithms to run, ["all"] to run all the available classifiers
-algos_monoview: ["decision_tree"]
+algos_monoview: ["decision_tree", "adaboost"]
# The names of the multiview algorithms to run, ["all"] to run all the available classifiers
-algos_multiview: ["weighted_linear_early_fusion", "weighted_linear_late_fusion",]
+algos_multiview: ["early_fusion_decision_tree", "early_fusion_adaboost", "weighted_linear_late_fusion",]
# The number of times the benchamrk is repeated with different train/test
# split, to have more statistically significant results
stats_iter: 1
diff --git a/summit/multiview_platform/monoview_classifiers/sgd.py b/summit/multiview_platform/monoview_classifiers/sgd.py
index 1b1d3375..312a2bf2 100644
--- a/summit/multiview_platform/monoview_classifiers/sgd.py
+++ b/summit/multiview_platform/monoview_classifiers/sgd.py
@@ -23,8 +23,8 @@ class SGD(SGDClassifier, BaseMonoviewClassifier):
loss=loss,
penalty=penalty,
alpha=alpha,
- max_iter=5,
- tol=None,
+ max_iter=max_iter,
+ tol=tol,
random_state=random_state
)
self.param_names = ["loss", "penalty", "alpha", "random_state"]
diff --git a/summit/multiview_platform/multiview_classifiers/additions/early_fusion_from_monoview.py b/summit/multiview_platform/multiview_classifiers/additions/early_fusion_from_monoview.py
new file mode 100644
index 00000000..582466b0
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/additions/early_fusion_from_monoview.py
@@ -0,0 +1,68 @@
+import numpy as np
+
+from ... import monoview_classifiers
+from ...multiview.multiview_utils import get_available_monoview_classifiers, \
+ BaseMultiviewClassifier, ConfigGenerator
+from ...utils.dataset import get_samples_views_indices
+from ...utils.multiclass import get_mc_estim, MultiClassWrapper
+
+# from ..utils.dataset import get_v
+
+classifier_class_name = "WeightedLinearEarlyFusion"
+
+
+class BaseEarlyFusion(BaseMultiviewClassifier):
+
+ def __init__(self, monoview_classifier="decision_tree", random_state=None,
+ **kwargs):
+ BaseMultiviewClassifier.__init__(self, random_state=random_state)
+ monoview_classifier_module = getattr(monoview_classifiers, monoview_classifier)
+ monoview_classifier_class = getattr(monoview_classifier_module, monoview_classifier_module.classifier_class_name)
+ self.monoview_classifier = monoview_classifier_class(**kwargs)
+
+ def set_params(self, **params):
+ self.monoview_classifier.set_params(**params)
+ return self
+
+ def get_params(self, deep=True):
+ monoview_params = self.monoview_classifier.get_params(deep=deep)
+ monoview_params["random_state"] = self.random_state
+ return monoview_params
+
+ def fit(self, X, y, train_indices=None, view_indices=None):
+ train_indices, X = self.transform_data_to_monoview(X, train_indices,
+ view_indices)
+ self.used_views = view_indices
+ if np.unique(y[train_indices]).shape[0] > 2 and \
+ not (isinstance(self.monoview_classifier, MultiClassWrapper)):
+ self.monoview_classifier = get_mc_estim(self.monoview_classifier,
+ self.random_state,
+ multiview=False,
+ y=y[train_indices])
+ self.monoview_classifier.fit(X, y[train_indices])
+ return self
+
+ def predict(self, X, sample_indices=None, view_indices=None):
+ _, X = self.transform_data_to_monoview(X, sample_indices, view_indices)
+ self._check_views(self.view_indices)
+ predicted_labels = self.monoview_classifier.predict(X)
+ return predicted_labels
+
+ def transform_data_to_monoview(self, dataset, sample_indices,
+ view_indices):
+ """Here, we extract the data from the HDF5 dataset file and store all
+ the concatenated views in one variable"""
+ sample_indices, self.view_indices = get_samples_views_indices(dataset,
+ sample_indices,
+ view_indices)
+
+ X = self.hdf5_to_monoview(dataset, sample_indices)
+ return sample_indices, X
+
+ def hdf5_to_monoview(self, dataset, samples):
+ """Here, we concatenate the views for the asked samples """
+ monoview_data = np.concatenate(
+ [dataset.get_v(view_idx, samples)
+ for index, view_idx
+ in enumerate(self.view_indices)], axis=1)
+ return monoview_data
\ No newline at end of file
diff --git a/summit/multiview_platform/multiview_classifiers/early_fusion_adaboost.py b/summit/multiview_platform/multiview_classifiers/early_fusion_adaboost.py
new file mode 100644
index 00000000..c35355f9
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/early_fusion_adaboost.py
@@ -0,0 +1,23 @@
+from .additions.early_fusion_from_monoview import BaseEarlyFusion
+from ..utils.hyper_parameter_search import CustomRandint
+from ..utils.base import base_boosting_estimators
+
+# from ..utils.dataset import get_v
+
+classifier_class_name = "EarlyFusionAdaboost"
+
+
+class EarlyFusionAdaboost(BaseEarlyFusion):
+
+ def __init__(self, random_state=None, n_estimators=50,
+ base_estimator=None, base_estimator_config=None, **kwargs):
+ BaseEarlyFusion.__init__(self, random_state=random_state,
+ monoview_classifier="adaboost",
+ n_estimators= n_estimators,
+ base_estimator=base_estimator,
+ base_estimator_config=base_estimator_config, **kwargs)
+ self.param_names = ["n_estimators", "base_estimator"]
+ self.classed_params = ["base_estimator"]
+ self.distribs = [CustomRandint(low=1, high=500),
+ base_boosting_estimators]
+ self.weird_strings = {"base_estimator": "class_name"}
\ No newline at end of file
diff --git a/summit/multiview_platform/multiview_classifiers/early_fusion_decision_tree.py b/summit/multiview_platform/multiview_classifiers/early_fusion_decision_tree.py
new file mode 100644
index 00000000..fc3e672b
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/early_fusion_decision_tree.py
@@ -0,0 +1,22 @@
+from .additions.early_fusion_from_monoview import BaseEarlyFusion
+from ..utils.hyper_parameter_search import CustomRandint
+
+# from ..utils.dataset import get_v
+
+classifier_class_name = "EarlyFusionDT"
+
+
+class EarlyFusionDT(BaseEarlyFusion):
+
+ def __init__(self, random_state=None, max_depth=None,
+ criterion='gini', splitter='best', **kwargs):
+ BaseEarlyFusion.__init__(self, random_state=random_state,
+ monoview_classifier="decision_tree", max_depth=max_depth,
+ criterion=criterion, splitter=splitter, **kwargs)
+ self.param_names = ["max_depth", "criterion", "splitter",
+ 'random_state']
+ self.classed_params = []
+ self.distribs = [CustomRandint(low=1, high=300),
+ ["gini", "entropy"],
+ ["best", "random"], [random_state]]
+ self.weird_strings = {}
\ No newline at end of file
diff --git a/summit/multiview_platform/multiview_classifiers/early_fusion_gradient_boosting.py b/summit/multiview_platform/multiview_classifiers/early_fusion_gradient_boosting.py
new file mode 100644
index 00000000..f87d3360
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/early_fusion_gradient_boosting.py
@@ -0,0 +1,21 @@
+from .additions.early_fusion_from_monoview import BaseEarlyFusion
+from ..utils.hyper_parameter_search import CustomRandint
+from ..monoview_classifiers.gradient_boosting import CustomDecisionTreeGB
+
+classifier_class_name = "EarlyFusionGB"
+
+
+class EarlyFusionGB(BaseEarlyFusion):
+
+ def __init__(self, random_state=None, loss="exponential", max_depth=1.0,
+ n_estimators=100,
+ init=CustomDecisionTreeGB(max_depth=1),
+ **kwargs):
+ BaseEarlyFusion.__init__(self, random_state=random_state,
+ monoview_classifier="gradient_boosting",
+ loss=loss, max_depth=max_depth,
+ n_estimators=n_estimators, init=init, **kwargs)
+ self.param_names = ["n_estimators", "max_depth"]
+ self.classed_params = []
+ self.distribs = [CustomRandint(low=50, high=500),
+ CustomRandint(low=1, high=10), ]
\ No newline at end of file
diff --git a/summit/multiview_platform/multiview_classifiers/early_fusion_lasso.py b/summit/multiview_platform/multiview_classifiers/early_fusion_lasso.py
new file mode 100644
index 00000000..828b7155
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/early_fusion_lasso.py
@@ -0,0 +1,17 @@
+from .additions.early_fusion_from_monoview import BaseEarlyFusion
+from ..utils.hyper_parameter_search import CustomUniform, CustomRandint
+
+classifier_class_name = "EarlyFusionLasso"
+
+
+class EarlyFusionLasso(BaseEarlyFusion):
+
+ def __init__(self, random_state=None, alpha=1.0,
+ max_iter=10, warm_start=False, **kwargs):
+ BaseEarlyFusion.__init__(self, random_state=None, alpha=alpha,
+ max_iter=max_iter,
+ warm_start=warm_start, **kwargs)
+ self.param_names = ["max_iter", "alpha", "random_state"]
+ self.classed_params = []
+ self.distribs = [CustomRandint(low=1, high=30--0),
+ CustomUniform(), [random_state]]
\ No newline at end of file
diff --git a/summit/multiview_platform/multiview_classifiers/early_fusion_random_forest.py b/summit/multiview_platform/multiview_classifiers/early_fusion_random_forest.py
new file mode 100644
index 00000000..1a01aa31
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/early_fusion_random_forest.py
@@ -0,0 +1,23 @@
+import numpy as np
+
+from .additions.early_fusion_from_monoview import BaseEarlyFusion
+from ..utils.hyper_parameter_search import CustomRandint
+
+classifier_class_name = "EarlyFusionRF"
+
+
+class EarlyFusionRF(BaseEarlyFusion):
+
+ def __init__(self, random_state=None, n_estimators=10,
+ max_depth=None, criterion='gini', **kwargs):
+ BaseEarlyFusion.__init__(self, random_state=random_state,
+ monoview_classifier="random_forest",
+ n_estimators=n_estimators, max_depth=max_depth,
+ criterion=criterion, **kwargs)
+ self.param_names = ["n_estimators", "max_depth", "criterion",
+ "random_state"]
+ self.classed_params = []
+ self.distribs = [CustomRandint(low=1, high=300),
+ CustomRandint(low=1, high=10),
+ ["gini", "entropy"], [random_state]]
+ self.weird_strings = {}
\ No newline at end of file
diff --git a/summit/multiview_platform/multiview_classifiers/early_fusion_sgd.py b/summit/multiview_platform/multiview_classifiers/early_fusion_sgd.py
new file mode 100644
index 00000000..580b4662
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/early_fusion_sgd.py
@@ -0,0 +1,19 @@
+from .additions.early_fusion_from_monoview import BaseEarlyFusion
+from ..utils.hyper_parameter_search import CustomUniform
+
+classifier_class_name = "EarlyFusionSGD"
+
+
+class EarlyFusionSGD(BaseEarlyFusion):
+
+ def __init__(self, random_state=None, loss='hinge',
+ penalty='l2', alpha=0.0001, max_iter=5, tol=None, **kwargs):
+ BaseEarlyFusion.__init__(self, random_state=random_state,
+ monoview_classifier="sgd", loss=loss,
+ penalty=penalty, alpha=alpha, max_iter=max_iter, tol=tol, **kwargs)
+ self.param_names = ["loss", "penalty", "alpha", "random_state"]
+ self.classed_params = []
+ self.distribs = [['log', 'modified_huber'],
+ ["l1", "l2", "elasticnet"],
+ CustomUniform(loc=0, state=1), [random_state]]
+ self.weird_strings = {}
\ No newline at end of file
diff --git a/summit/multiview_platform/multiview_classifiers/early_fusion_svm_rbf.py b/summit/multiview_platform/multiview_classifiers/early_fusion_svm_rbf.py
new file mode 100644
index 00000000..6d427c98
--- /dev/null
+++ b/summit/multiview_platform/multiview_classifiers/early_fusion_svm_rbf.py
@@ -0,0 +1,13 @@
+from .additions.early_fusion_from_monoview import BaseEarlyFusion
+from ..utils.hyper_parameter_search import CustomUniform
+
+classifier_class_name = "EarlyFusionSVMRBF"
+
+
+class EarlyFusionSVMRBF(BaseEarlyFusion):
+
+ def __init__(self, random_state=None, C=1.0, **kwargs):
+ BaseEarlyFusion.__init__(self, random_state=random_state,
+ monoview_classifier="svm_rbf", C=C, **kwargs)
+ self.param_names = ["C", "random_state"]
+ self.distribs = [CustomUniform(loc=0, state=1), [random_state]]
\ No newline at end of file
--
GitLab