Fix typos (scikit-learn#6942)

Author: shotat

benchmarks/bench_plot_randomized_svd.py

@@ -373,7 +373,7 @@ def bench_b(power_list):
     if enable_spectral_norm:
         title = "%s: spectral norm diff vs n power iteration" % (dataset_name)
         plot_power_iter_vs_s(power_iter, all_spectral, title)
-    title = "%s: frobenius norm diff vs n power iteration" % (dataset_name)
+    title = "%s: Frobenius norm diff vs n power iteration" % (dataset_name)
     plot_power_iter_vs_s(power_iter, all_frobenius, title)
 
 

examples/calibration/plot_calibration_multiclass.py

@@ -145,7 +145,7 @@ class of an instance (red: class 1, green: class 2, blue: class 3).
                         zip(calibrated_classifier.calibrators_, p.T)]).T
 prediction /= prediction.sum(axis=1)[:, None]
 
-# Ploit modifications of calibrator
+# Plot modifications of calibrator
 for i in range(prediction.shape[0]):
     plt.arrow(p[i, 0], p[i, 1],
               prediction[i, 0] - p[i, 0], prediction[i, 1] - p[i, 1],

examples/ensemble/plot_ensemble_oob.py

@@ -41,7 +41,7 @@
                            random_state=RANDOM_STATE)
 
 # NOTE: Setting the `warm_start` construction parameter to `True` disables
-# support for paralellised ensembles but is necessary for tracking the OOB
+# support for parallelized ensembles but is necessary for tracking the OOB
 # error trajectory during training.
 ensemble_clfs = [
     ("RandomForestClassifier, max_features='sqrt'",

examples/hetero_feature_union.py

@@ -165,7 +165,7 @@ def transform(self, posts):
     ('svc', SVC(kernel='linear')),
 ])
 
-# limit the list of categories to make running this exmaple faster.
+# limit the list of categories to make running this example faster.
 categories = ['alt.atheism', 'talk.religion.misc']
 train = fetch_20newsgroups(random_state=1,
                            subset='train',

examples/linear_model/plot_ard.py

@@ -39,7 +39,7 @@
 relevant_features = np.random.randint(0, n_features, 10)
 for i in relevant_features:
     w[i] = stats.norm.rvs(loc=0, scale=1. / np.sqrt(lambda_))
-# Create noite with a precision alpha of 50.
+# Create noise with a precision alpha of 50.
 alpha_ = 50.
 noise = stats.norm.rvs(loc=0, scale=1. / np.sqrt(alpha_), size=n_samples)
 # Create the target

examples/linear_model/plot_lasso_coordinate_descent_path.py

@@ -40,7 +40,7 @@
 alphas_enet, coefs_enet, _ = enet_path(
     X, y, eps=eps, l1_ratio=0.8, fit_intercept=False)
 
-print("Computing regularization path using the positve elastic net...")
+print("Computing regularization path using the positive elastic net...")
 alphas_positive_enet, coefs_positive_enet, _ = enet_path(
     X, y, eps=eps, l1_ratio=0.8, positive=True, fit_intercept=False)
 

sklearn/linear_model/randomized_l1.py

@@ -546,7 +546,7 @@ def _lasso_stability_path(X, y, mask, weights, eps):
                                      alpha_min=alpha_min)
     # Scale alpha by alpha_max
     alphas /= alphas[0]
-    # Sort alphas in assending order
+    # Sort alphas in ascending order
     alphas = alphas[::-1]
     coefs = coefs[:, ::-1]
     # Get rid of the alphas that are too small
@@ -564,7 +564,7 @@ def lasso_stability_path(X, y, scaling=0.5, random_state=None,
                          sample_fraction=0.75,
                          eps=4 * np.finfo(np.float).eps, n_jobs=1,
                          verbose=False):
-    """Stabiliy path based on randomized Lasso estimates
+    """Stability path based on randomized Lasso estimates
 
     Read more in the :ref:`User Guide <randomized_l1>`.
 

sklearn/linear_model/sag.py

@@ -110,10 +110,10 @@ def sag_solver(X, y, sample_weight=None, loss='log', alpha=1.,
 
     max_iter: int, optional
         The max number of passes over the training data if the stopping
-        criterea is not reached. Defaults to 1000.
+        criteria is not reached. Defaults to 1000.
 
     tol: double, optional
-        The stopping criterea for the weights. The iterations will stop when
+        The stopping criteria for the weights. The iterations will stop when
         max(change in weights) / max(weights) < tol. Defaults to .001
 
     verbose: integer, optional

sklearn/linear_model/sag_fast.pyx

@@ -264,9 +264,9 @@ def sag(SequentialDataset dataset,
     # the index (row number) of the current sample
     cdef int sample_ind
 
-    # the maximum change in weights, used to compute stopping criterea
+    # the maximum change in weights, used to compute stopping criteria
     cdef double max_change
-    # a holder variable for the max weight, used to compute stopping criterea
+    # a holder variable for the max weight, used to compute stopping criteria
     cdef double max_weight
 
     # the start time of the fit

sklearn/linear_model/sgd_fast.pyx

@@ -482,8 +482,8 @@ def average_sgd(np.ndarray[double, ndim=1, mode='c'] weights,
         (1) constant, eta = eta0
         (2) optimal, eta = 1.0/(alpha * t).
         (3) inverse scaling, eta = eta0 / pow(t, power_t)
-        (4) Passive Agressive-I, eta = min(alpha, loss/norm(x))
-        (5) Passive Agressive-II, eta = 1.0 / (norm(x) + 0.5*alpha)
+        (4) Passive Aggressive-I, eta = min(alpha, loss/norm(x))
+        (5) Passive Aggressive-II, eta = 1.0 / (norm(x) + 0.5*alpha)
     eta0 : double
         The initial learning rate.
     power_t : double
@@ -503,9 +503,9 @@ def average_sgd(np.ndarray[double, ndim=1, mode='c'] weights,
     intercept : float
         The fitted intercept term.
     average_weights : array shape=[n_features]
-        The averaged weights accross iterations
+        The averaged weights across iterations
     average_intercept : float
-        The averaged intercept accross iterations
+        The averaged intercept across iterations
     """
     return _plain_sgd(weights,
                       intercept,