Skip to content

Instantly share code, notes, and snippets.

@pprett

pprett/linearsvc_vs_svc.py

Created May 24, 2011 12:00
Show Gist options
  • Save pprett/988586 to your computer and use it in GitHub Desktop.
Save pprett/988586 to your computer and use it in GitHub Desktop.
Download ZIP

Revisions

  1. pprett revised this gist May 24, 2011. 1 changed file with 5 additions and 4 deletions.
    9 changes: 5 additions & 4 deletions linearsvc_vs_svc.py
    View file
    Original file line number Diff line number Diff line change
    @@ -14,6 +14,7 @@
    from scikits.learn.grid_search import GridSearchCV
    from scikits.learn.metrics.metrics import f1_score
    from scikits.learn.cross_val import StratifiedKFold
    from scikits.learn.preprocessing import Scaler

    # Initialize default C and gamma values
    C_start, C_end, C_step = -3, 4, 2
    @@ -33,10 +34,10 @@
    train, test = iter(StratifiedKFold(Y, 2, indices=True)).next()

    # standardize data - try to comment this out to see the effect!
    mean, std = X[train].mean(axis=0), X[train].std(axis=0)
    std[std == 0.0] = 1.0
    X[train] = (X[train] - mean) / std
    X[test] = (X[test] - mean) / std
    scaler = Scaler()
    scaler.fit(X[train])
    X[train] = scaler.transform(X[train], copy=False)
    X[test] = scaler.transform(X[test], copy=False)

    # make X sparse
    X = sparse.csr_matrix(X)
  2. pprett created this gist May 24, 2011.
    79 changes: 79 additions & 0 deletions linearsvc_vs_svc.py
    View file
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,79 @@
    """High difference in classifier accuracies with LinearSVC and SVC.
    Get data.npz from [1].
    [1] https://docs.google.com/leaf?id=0B1BhwRZOwyxRZTcxZDA1OWMtZjZkMy00YjgxLWI3ZTMtZjJkNGIyODAyOTQy&hl=en_US
    """
    print __doc__

    import numpy as np
    from functools import partial

    from scipy import sparse
    from scikits.learn import svm
    from scikits.learn.grid_search import GridSearchCV
    from scikits.learn.metrics.metrics import f1_score
    from scikits.learn.cross_val import StratifiedKFold

    # Initialize default C and gamma values
    C_start, C_end, C_step = -3, 4, 2


    if __name__ == "__main__":
    cross_fold = 10

    A = np.load("data.npz")

    Y = A["arr_1"]
    X = A["arr_0"]
    print "X.shape=", X.shape
    print "Y.shape=", Y.shape

    folds = StratifiedKFold(Y, cross_fold, indices=True)
    train, test = iter(StratifiedKFold(Y, 2, indices=True)).next()

    # standardize data - try to comment this out to see the effect!
    mean, std = X[train].mean(axis=0), X[train].std(axis=0)
    std[std == 0.0] = 1.0
    X[train] = (X[train] - mean) / std
    X[test] = (X[test] - mean) / std

    # make X sparse
    X = sparse.csr_matrix(X)

    # Generate grid search values for C, gamma
    C_val = 2. ** np.arange(C_start, C_end + C_step, C_step)
    tol_val = [0.1, 0.01, 0.001, 0.0001]
    params = {'C': C_val, 'tol': tol_val}

    for clf_class in [svm.sparse.LinearSVC, partial(svm.sparse.SVC,
    kernel="linear")]:
    grid_clf = clf_class()
    print "_" * 80
    print grid_clf
    print

    grid_search = GridSearchCV(grid_clf, params, score_func=f1_score)
    grid_search.fit(X[train], Y[train],
    cv=StratifiedKFold(Y[train],
    10, indices=True))
    y_true, y_pred = Y[test], grid_search.predict(X[test])

    print "Classification report for the best estimator: "
    print grid_search.best_estimator

    print "Tuned for with optimal f1-score: %0.3f" % f1_score(y_true,
    y_pred)

    print "Best score: %0.3f" % grid_search.best_score

    best_parameters = grid_search.best_estimator._get_params()
    print "Best C: %0.3f " % best_parameters['C']
    print "Best tolerance: %0.16f " % best_parameters['tol']

    clf = clf_class(C=best_parameters['C'], tol=best_parameters['tol'])
    print clf
    clf.fit(X[train], Y[train])
    y_pred = clf.predict(X[test])
    print "Accuracy:\t%.4f" % (y_true == y_pred).mean()
    print "F-Score:\t%.4f" % f1_score(y_true, y_pred)