diff --git a/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/additions/kernel_learning.py b/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/additions/kernel_learning.py
index 3eda3730f720c90403aff91ee3aa3757eae032e9..c2a18ed33c066a15229c60867fee3145b6ec081b 100644
--- a/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/additions/kernel_learning.py
+++ b/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/additions/kernel_learning.py
@@ -1,8 +1,9 @@
from sklearn.metrics import pairwise
import numpy as np
-from ...multiview.multiview_utils import BaseMultiviewClassifier
+from ...multiview.multiview_utils import BaseMultiviewClassifier, get_examples_views_indices
from ...utils.hyper_parameter_search import CustomUniform, CustomRandint
+from ...utils.transformations import sign_labels, unsign_labels
class KernelClassifier(BaseMultiviewClassifier):
@@ -21,6 +22,23 @@ class KernelClassifier(BaseMultiviewClassifier):
**kernel_config)
return new_X
+ def _init_fit(self, X, y, train_indices, view_indices):
+ train_indices, view_indices = get_examples_views_indices(X,
+ train_indices,
+ view_indices)
+ self.init_kernels(nb_view=len(view_indices), )
+ new_X = self._compute_kernels(X,
+ train_indices, view_indices)
+ new_y = sign_labels(y[train_indices])
+ return new_X, new_y
+
+ def extract_labels(self, predicted_labels):
+ signed_labels = np.sign(predicted_labels)
+ return unsign_labels(signed_labels)
+
+
+
+
def init_kernels(self, nb_view=2, ):
if isinstance(self.kernel_types, KernelDistribution):
self.kernel_functions = self.kernel_types.draw(nb_view)
@@ -58,11 +76,11 @@ class KernelConfigDistribution:
def __init__(self, seed=42):
self.random_state=np.random.RandomState(seed)
self.possible_config = {
- "polynomial_kernel":{"degree": CustomRandint(low=1, high=7),
- "gamma": CustomUniform(),
- "coef0": CustomUniform()
-
- },
+ # "polynomial_kernel":{"degree": CustomRandint(low=1, high=7),
+ # "gamma": CustomUniform(),
+ # "coef0": CustomUniform()
+ #
+ # },
"chi2_kernel": {"gamma": CustomUniform()},
"rbf_kernel": {"gamma": CustomUniform()},
}
@@ -90,8 +108,7 @@ class KernelDistribution:
def __init__(self, seed=42):
self.random_state=np.random.RandomState(seed)
- self.available_kernels = [pairwise.polynomial_kernel,
- pairwise.chi2_kernel,
+ self.available_kernels = [pairwise.chi2_kernel,
pairwise.rbf_kernel,]
def draw(self, nb_view):
diff --git a/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/lp_norm_mkl.py b/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/lp_norm_mkl.py
index f68973f8eb61546c2a7d0f58d783c009fe06366f..f93f632c0982f7203f02feb727095dea6a1b966b 100644
--- a/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/lp_norm_mkl.py
+++ b/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/lp_norm_mkl.py
@@ -147,7 +147,6 @@ class MKL(BaseEstimator, ClassifierMixin):
else:
return C, weights
-
def predict(self, X, views_ind=None):
if isinstance(X, Metriclearn_array):
# self.X_ = X
@@ -157,10 +156,9 @@ class MKL(BaseEstimator, ClassifierMixin):
elif isinstance(X, np.ndarray):
X = Metriclearn_array(X, views_ind)
C = self.C
- weights = self.weights
+ weights = self.weights
return self.lpMKL_predict(X , C, weights)
-
def lpMKL_predict(self, X, C, weights, views_ind=None):
if isinstance(X, Metriclearn_array):
# self.X_ = X
@@ -177,7 +175,6 @@ class MKL(BaseEstimator, ClassifierMixin):
# kernel = weights[0] * self.data.test_kernel_dict[0]
# for v in range(1, views):
# kernel = kernel + weights[v] * self.data.test_kernel_dict[v]
-
# TEST KERNEL APPROXIMATION
kernel = np.zeros((tt, self.X_.shape[0]))
for v in range(0, views):
@@ -221,19 +218,15 @@ class LPNormMKL(KernelClassifier, MKL):
self.prev_alpha = prev_alpha
def fit(self, X, y, train_indices=None, view_indices=None):
- train_indices, view_indices = get_examples_views_indices(X, train_indices,
- view_indices)
- self.init_kernels(nb_view= len(view_indices), )
- new_X = self._compute_kernels(X,
- train_indices, view_indices)
- return super(LPNormMKL, self).fit(new_X, y[train_indices])
+ new_X, new_y = self._init_fit(X, y, train_indices, view_indices)
+ return super(LPNormMKL, self).fit(new_X, new_y)
def predict(self, X, example_indices=None, view_indices=None):
example_indices, view_indices = get_examples_views_indices(X,
example_indices,
view_indices)
new_X = self._compute_kernels(X, example_indices, view_indices)
- return super(LPNormMKL, self).predict(new_X)
+ return self.extract_labels(super(LPNormMKL, self).predict(new_X))
diff --git a/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/mvml.py b/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/mvml.py
index 5054ec54eb565f31488c29d4f312fafc2b03133b..3e5af0a9cc83bf1ef65fb1a027d23efb57c97321 100644
--- a/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/mvml.py
+++ b/multiview_platform/mono_multi_view_classifiers/multiview_classifiers/mvml.py
@@ -503,17 +503,14 @@ class MVMLClassifier(KernelClassifier, MVML):
[0,1], CustomRandint(low=1, high=100), [[0.1,0.9]]]
def fit(self, X, y, train_indices=None, view_indices=None):
- train_indices, view_indices = get_examples_views_indices(X,
- train_indices,
- view_indices)
- self.init_kernels(nb_view=len(view_indices), )
- new_X = self._compute_kernels(X,
- train_indices, view_indices)
- return super(MVMLClassifier, self).fit(new_X, y[train_indices])
+ new_X, new_y = self._init_fit(X, y, train_indices, view_indices)
+ return super(MVMLClassifier, self).fit(new_X, new_y)
def predict(self, X, example_indices=None, view_indices=None):
example_indices, view_indices = get_examples_views_indices(X,
example_indices,
view_indices)
new_X = self._compute_kernels(X, example_indices, view_indices)
- return super(MVMLClassifier, self).predict(new_X)
\ No newline at end of file
+ print(self.extract_labels(super(MVMLClassifier, self).predict(new_X)))
+ return self.extract_labels(super(MVMLClassifier, self).predict(new_X))
+
diff --git a/multiview_platform/mono_multi_view_classifiers/utils/transformations.py b/multiview_platform/mono_multi_view_classifiers/utils/transformations.py
index 9d26ddde8bd02fea2ef1176385ac251260027e40..e2cdfdf132364ec0da54d98591d202b116dda339 100644
--- a/multiview_platform/mono_multi_view_classifiers/utils/transformations.py
+++ b/multiview_platform/mono_multi_view_classifiers/utils/transformations.py
@@ -1,8 +1,45 @@
import numpy as np
-
+from sklearn.preprocessing import LabelEncoder
def sign_labels(labels):
- if set(labels) == (0, 1):
+ """
+ Returns a label array with (-1,1) as labels.
+ If labels was already made of (-1,1), returns labels.
+ If labels is made of (0,1), returns labels with all
+ zeros transformed in -1.
+
+ Parameters
+ ----------
+ labels
+
+ The original label numpy array
+
+ Returns
+ -------
+ A np.array with labels made of (-1,1)
+ """
+ if 0 in labels:
return np.array([label if label != 0 else -1 for label in labels])
else:
return labels
+
+def unsign_labels(labels):
+ """
+ The inverse function
+
+ Parameters
+ ----------
+ labels
+
+ Returns
+ -------
+
+ """
+ if len(labels.shape)==2:
+ labels = labels.reshape((labels.shape[0], ))
+ print(labels)
+ if -1 in labels:
+ return np.array([label if label != -1 else 0 for label in labels])
+ else:
+ return labels
+