Skip to content

Updated LabelBinarizer and cross_val_predict #11

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
isshamie wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Updated LabelBinarizer and cross_val_predict #11

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion pyopls/permutation_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -314,7 +314,12 @@ def _permutation_test_score(estimator, X, y, groups=None, cv='warn',
"""Auxiliary function for permutation_test_score"""
if score_functions is None:
score_functions = [r2_score]
y_pred = cross_val_predict(estimator, X, y, groups, cv, n_jobs, verbose, fit_params, pre_dispatch, method)
try:
y_pred = cross_val_predict(estimator, X, y, groups=groups, cv=cv, n_jobs=n_jobs,
verbose=verbose, fit_params=fit_params, pre_dispatch=pre_dispatch, method=method)
except TypeError:
y_pred = cross_val_predict(estimator, X, y, groups=groups, cv=cv, n_jobs=n_jobs,
verbose=verbose, params=fit_params, pre_dispatch=pre_dispatch, method=method)
cv_scores = [score_function(y, y_pred) for score_function in score_functions]
return np.array(cv_scores)

Expand Down
12 changes: 6 additions & 6 deletions pyopls/validation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -205,12 +205,12 @@ def _get_score_function(self, Y):
def _validate(self, X, Y, n_components, score_function, cv=None, n_jobs=None, verbose=0, pre_dispatch='2*n_jobs'):
cv = cv or self._get_validator(Y, self.k)
Z = OPLS(n_components, self.scale).fit_transform(X, Y)
y_pred = cross_val_predict(PLSRegression(1, self.scale), Z, Y, cv=cv, n_jobs=n_jobs, verbose=verbose,
y_pred = cross_val_predict(PLSRegression(1, scale=self.scale), Z, Y, cv=cv, n_jobs=n_jobs, verbose=verbose,
pre_dispatch=pre_dispatch)
return score_function(Y, y_pred)

def _process_binary_target(self, y, pos_label=None):
self.binarizer_ = LabelBinarizer(-1, 1)
self.binarizer_ = LabelBinarizer(neg_label=-1, pos_label=1)
self.binarizer_.fit(y)
if pos_label is not None and self.binarizer_.transform([pos_label])[0] == -1:
self.binarizer_.classes_ = np.flip(self.binarizer_.classes_)
Expand Down Expand Up @@ -367,7 +367,7 @@ def _determine_significant_features(self,
"""
Z = OPLS(n_components, self.scale).fit_transform(X, y)
x_loadings, permutation_x_loadings, p_values = feature_permutation_loading(
PLSRegression(1, self.scale), Z, y, self.n_inner_permutations, self.inner_alpha,
PLSRegression(1, scale=self.scale), Z, y, self.n_inner_permutations, self.inner_alpha,
self.n_outer_permutations, random_state, n_jobs, verbose, pre_dispatch
)
return p_values < self.outer_alpha, p_values, permutation_x_loadings
Expand All @@ -376,7 +376,7 @@ def cross_val_roc_curve(self, X, y, cv=None, n_jobs=None, verbose=0, pre_dispatc
Z = self.opls_.transform(X)
cv = cv or self._get_validator(y, self.k)
check_is_fitted(self, ['opls_', 'pls_', 'binarizer_'])
y_pred = cross_val_predict(PLSRegression(1, self.scale), Z, y, cv=cv, n_jobs=n_jobs, verbose=verbose,
y_pred = cross_val_predict(PLSRegression(1, scale=self.scale), Z, y, cv=cv, n_jobs=n_jobs, verbose=verbose,
pre_dispatch=pre_dispatch)
return roc_curve(y, y_pred)

Expand Down Expand Up @@ -462,7 +462,7 @@ def _log(txt):

self.opls_ = OPLS(self.n_components_, self.scale).fit(X, y)
Z = self.opls_.transform(X)
self.pls_ = PLSRegression(1, self.scale).fit(Z, y)
self.pls_ = PLSRegression(1, scale=self.scale).fit(Z, y)
self.r_squared_X_ = self.opls_.score(X)
y_pred = self.pls_.predict(Z)
self.r_squared_Y_ = r2_score(y, y_pred)
Expand All @@ -477,7 +477,7 @@ def _log(txt):

_log('Performing cross-validated metric permutation tests.')

cv_results = permutation_test_score(PLSRegression(1, self.scale), Z, y, cv=cv,
cv_results = permutation_test_score(PLSRegression(1, scale=self.scale), Z, y, cv=cv,
n_permutations=self.n_permutations, cv_score_functions=score_functions,
n_jobs=n_jobs, verbose=verbose, pre_dispatch=pre_dispatch)
if self.is_discrimination(y):
Expand Down