8000 Typos found by codespell · scikit-learn/scikit-learn@50883ae · GitHub
[go: up one dir, main page]
Skip to content

Commit 50883ae

Browse files
DimitriPapadopoulosDimitriPapadopoulos
committed
Typos found by codespell
1 parent 391dbe6 commit 50883ae

File tree

124 files changed

+175
-175
lines changed
  • _ransac.py
  • _ridge.py
  • _sag_fast.pyx.tp
  • _stochastic_gradient.py
  • tests
  • manifold
  • metrics
  • model_selection
  • neighbors
  • neural_network/tests
  • preprocessing
  • semi_supervised
  • svm/src/libsvm
  • tree
  • utils

    • Some content is hidden

    Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

    124 files changed

    +175
    -175
    lines changed

    ‎benchmarks/bench_mnist.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -6,7 +6,7 @@
    66
    Benchmark on the MNIST dataset. The dataset comprises 70,000 samples
    77
    and 784 features. Here, we consider the task of predicting
    88
    10 classes - digits from 0 to 9 from their raw images. By contrast to the
    9-
    covertype dataset, the feature space is homogenous.
    9+
    covertype dataset, the feature space is homogeneous.
    1010
    1111
    Example of output :
    1212
    [..]

    ‎benchmarks/bench_random_projections.py

    Lines changed: 3 additions & 3 deletions
    Original file line numberDiff line numberDiff line change
    @@ -43,10 +43,10 @@ def compute_time(t_start, delta):
    4343
    return delta.seconds + delta.microseconds / mu_second
    4444

    4545

    46-
    def bench_scikit_transformer(X, transfomer):
    46+
    def bench_scikit_transformer(X, transformer):
    4747
    gc.collect()
    4848

    49-
    clf = clone(transfomer)
    49+
    clf = clone(transformer)
    5050

    5151
    # start time
    5252
    t_start = datetime.now()
    @@ -195,7 +195,7 @@ def print_row(clf_type, time_fit, time_transform):
    195195
    ###########################################################################
    196196
    n_nonzeros = int(opts.ratio_nonzeros * opts.n_features)
    197197

    198-
    print("Dataset statics")
    198+
    print("Dataset statistics")
    199199
    print("===========================")
    200200
    print("n_samples \t= %s" % opts.n_samples)
    201201
    print("n_features \t= %s" % opts.n_features)

    ‎build_tools/azure/posix-docker.yml

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -39,7 +39,7 @@ jobs:
    3939
    ${{ insert }}: ${{ parameters.matrix }}
    4040

    4141
    steps:
    42-
    # Container is detached and sleeping, allowing steps to run commmands
    42+
    # Container is detached and sleeping, allowing steps to run commands
    4343
    # in the container. The TEST_DIR is mapped allowing the host to access
    4444
    # the JUNITXML file
    4545
    - script: >

    ‎build_tools/circle/list_versions.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -34,7 +34,7 @@ def human_readable_data_quantity(quantity, multiple=1024):
    3434

    3535
    def get_file_extension(version):
    3636
    if "dev" in version:
    37-
    # The 'dev' branch should be explictly handled
    37+
    # The 'dev' branch should be explicitly handled
    3838
    return "zip"
    3939

    4040
    current_version = LooseVersion(version)

    ‎build_tools/shared.sh

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -5,7 +5,7 @@ get_dep() {
    55
    # do not install with none
    66
    echo
    77
    elif [[ "${version%%[^0-9.]*}" ]]; then
    8-
    # version number is explicity passed
    8+
    # version number is explicitly passed
    99
    echo "$package==$version"
    1010
    elif [[ "$version" == "latest" ]]; then
    1111
    # use latest

    ‎doc/common_pitfalls.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -560,7 +560,7 @@ bad performance. Similarly, we want a random forest to be robust w.r.t the
    560560
    set of randomly selected features that each tree will be using.
    561561

    562562
    For these reasons, it is preferable to evaluate the cross-validation
    563-
    preformance by letting the estimator use a different RNG on each fold. This
    563+
    performance by letting the estimator use a different RNG on each fold. This
    564564
    is done by passing a `RandomState` instance (or `None`) to the estimator
    565565
    initialization.
    566566

    ‎doc/conf.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -240,7 +240,7 @@
    240240
    "release_highlights"
    241241
    ] = f"auto_examples/release_highlights/{latest_highlights}"
    242242

    243-
    # get version from higlight name assuming highlights have the form
    243+
    # get version from highlight name assuming highlights have the form
    244244
    # plot_release_highlights_0_22_0
    245245
    highlight_version = ".".join(latest_highlights.split("_")[-3:-1])
    246246
    html_context["release_highlights_version"] = highlight_version

    ‎doc/developers/advanced_installation.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -377,7 +377,7 @@ isolation from the Python packages installed via the system packager. When
    377377
    using an isolated environment, ``pip3`` should be replaced by ``pip`` in the
    378378
    above commands.
    379379

    380-
    When precompiled wheels of the runtime dependencies are not avalaible for your
    380+
    When precompiled wheels of the runtime dependencies are not available for your
    381381
    architecture (e.g. ARM), you can install the system versions:
    382382

    383383
    .. prompt:: bash $

    ‎doc/developers/contributing.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -1004,7 +1004,7 @@ installed in your current Python environment:
    10041004

    10051005
    asv run --python=same
    10061006

    1007-
    It's particulary useful when you installed scikit-learn in editable mode to
    1007+
    It's particularly useful when you installed scikit-learn in editable mode to
    10081008
    avoid creating a new environment each time you run the benchmarks. By default
    10091009
    the results are not saved when using an existing installation. To save the
    10101010
    results you must specify a commit hash:

    ‎doc/developers/maintainer.rst

    Lines changed: 2 additions & 2 deletions
    Original file line numberDiff line numberDiff line change
    @@ -33,7 +33,7 @@ Before a release
    3333

    3434
    - ``maint_tools/sort_whats_new.py`` can put what's new entries into
    3535
    sections. It's not perfect, and requires manual checking of the changes.
    36-
    If the whats new list is well curated, it may not be necessary.
    36+
    If the what's new list is well curated, it may not be necessary.
    3737

    3838
    - The ``maint_tools/whats_missing.sh`` script may be used to identify pull
    3939
    requests that were merged but likely missing from What's New.
    @@ -198,7 +198,7 @@ Making a release
    198198
    `Continuous Integration
    199199
    <https://en.wikipedia.org/wiki/Continuous_integration>`_. The CD workflow on
    200200
    GitHub Actions is also used to automatically create nightly builds and
    201-
    publish packages for the developement branch of scikit-learn. See
    201+
    publish packages for the development branch of scikit-learn. See
    202202
    :ref:`install_nightly_builds`.
    203203

    204204
    4. Once all the CD jobs have completed successfully in the PR, merge it,

    ‎doc/install.rst

    Lines changed: 2 additions & 2 deletions
    Original file line numberDiff line numberDiff line change
    @@ -158,7 +158,7 @@ Installing on Apple Silicon M1 hardware
    158158

    159159
    The r F438 ecently introduced `macos/arm64` platform (sometimes also known as
    160160
    `macos/aarch64`) requires the open source community to upgrade the build
    161-
    configuation and automation to properly support it.
    161+
    configuration and automation to properly support it.
    162162

    163163
    At the time of writing (January 2021), the only way to get a working
    164164
    installation of scikit-learn on this hardware is to install scikit-learn and its
    @@ -204,7 +204,7 @@ It can be installed by typing the following command:
    204204
    Debian/Ubuntu
    205205
    -------------
    206206

    207-
    The Debian/Ubuntu package is splitted in three different packages called
    207+
    The Debian/Ubuntu package is split in three different packages called
    208208
    ``python3-sklearn`` (python modules), ``python3-sklearn-lib`` (low-level
    209209
    implementations and bindings), ``python3-sklearn-doc`` (documentation).
    210210
    Only the Python 3 version is available in the Debian Buster (the more recent

    ‎doc/modules/compose.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -573,7 +573,7 @@ many estimators. This visualization is activated by setting the
    573573

    574574
    >>> from sklearn import set_config
    575575
    >>> set_config(display='diagram') # doctest: +SKIP
    576-
    >>> # diplays HTML representation in a jupyter context
    576+
    >>> # displays HTML representation in a jupyter context
    577577
    >>> column_trans # doctest: +SKIP
    578578

    579579
    An example of the HTML output can be seen in the

    ‎doc/modules/cross_decomposition.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -64,7 +64,7 @@ Set :math:`X_1` to :math:`X` and :math:`Y_1` to :math:`Y`. Then, for each
    6464
    :math:`C = X_k^T Y_k`.
    6565
    :math:`u_k` and :math:`v_k` are called the *weights*.
    6666
    By definition, :math:`u_k` and :math:`v_k` are
    67-
    choosen so that they maximize the covariance between the projected
    67+
    chosen so that they maximize the covariance between the projected
    6868
    :math:`X_k` and the projected target, that is :math:`\text{Cov}(X_k u_k,
    6969
    Y_k v_k)`.
    7070
    - b) Project :math:`X_k` and :math:`Y_k` on the singular vectors to obtain

    ‎doc/modules/cross_validation.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -974,7 +974,7 @@ test is therefore only able to show when the model reliably outperforms
    974974
    random guessing.
    975975

    976976
    Finally, :func:`~sklearn.model_selection.permutation_test_score` is computed
    977-
    using brute force and interally fits ``(n_permutations + 1) * n_cv`` models.
    977+
    using brute force and internally fits ``(n_permutations + 1) * n_cv`` models.
    978978
    It is therefore only tractable with small datasets for which fitting an
    979979
    individual model is very fast.
    980980

    ‎doc/modules/decomposition.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -829,7 +829,7 @@ and the intensity of the regularization with the :attr:`alpha_W` and :attr:`alph
    829829
    (:math:`\alpha_W` and :math:`\alpha_H`) parameters. The priors are scaled by the number
    830830
    of samples (:math:`n\_samples`) for `H` and the number of features (:math:`n\_features`)
    831831
    for `W` to keep their impact balanced with respect to one another and to the data fit
    832-
    term as independant as possible of the size of the training set. Then the priors terms
    832+
    term as independent as possible of the size of the training set. Then the priors terms
    833833
    are:
    834834

    835835
    .. math::

    ‎doc/modules/lda_qda.rst

    Lines changed: 2 additions & 2 deletions
    Original file line numberDiff line numberDiff line change
    @@ -187,7 +187,7 @@ an estimate for the covariance matrix). Setting this parameter to a value
    187187
    between these two extrema will estimate a shrunk version of the covariance
    188188
    matrix.
    189189

    190-
    The shrinked Ledoit and Wolf estimator of covariance may not always be the
    190+
    The shrunk Ledoit and Wolf estimator of covariance may not always be the
    191191
    best choice. For example if the distribution of the data
    192192
    is normally distributed, the
    193193
    Oracle Shrinkage Approximating estimator :class:`sklearn.covariance.OAS`
    @@ -234,7 +234,7 @@ For QDA, the use of the SVD solver relies on the fact that the covariance
    234234
    matrix :math:`\Sigma_k` is, by definition, equal to :math:`\frac{1}{n - 1}
    235235
    X_k^tX_k = \frac{1}{n - 1} V S^2 V^t` where :math:`V` comes from the SVD of the (centered)
    236236
    matrix: :math:`X_k = U S V^t`. It turns out that we can compute the
    237-
    log-posterior above without having to explictly compute :math:`\Sigma`:
    237+
    log-posterior above without having to explicitly compute :math:`\Sigma`:
    238238
    computing :math:`S` and :math:`V` via the SVD of :math:`X` is enough. For
    239239
    LDA, two SVDs are computed: the SVD of the centered input matrix :math:`X`
    240240
    and the SVD of the class-wise mean vectors.

    ‎doc/modules/model_evaluation.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -2381,7 +2381,7 @@ of 0.0.
    23812381
    A scorer object with a specific choice of ``power`` can be built by::
    23822382

    23832383
    >>> from sklearn.metrics import d2_tweedie_score, make_scorer
    2384-
    >>> d2_tweedie_score_15 = make_scorer(d2_tweedie_score, pwoer=1.5)
    2384+
    >>> d2_tweedie_score_15 = make_scorer(d2_tweedie_score, power=1.5)
    23852385

    23862386
    .. _pinball_loss:
    23872387

    ‎doc/modules/outlier_detection.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -332,7 +332,7 @@ chosen 1) greater than the minimum number of objects a cluster has to contain,
    332332
    so that other objects can be local outliers relative to this cluster, and 2)
    333333
    smaller than the maximum number of close by objects that can potentially be
    334334
    local outliers.
    335-
    In practice, such informations are generally not available, and taking
    335+
    In practice, such information are generally not available, and taking
    336336
    n_neighbors=20 appears to work well in general.
    337337
    When the proportion of outliers is high (i.e. greater than 10 \%, as in the
    338338
    example below), n_neighbors should be greater (n_neighbors=35 in the example

    ‎doc/modules/sgd.rst

    Lines changed: 2 additions & 2 deletions
    Original file line numberDiff line numberDiff line change
    @@ -123,7 +123,7 @@ Please refer to the :ref:`mathematical section below
    123123
    The first two loss functions are lazy, they only update the model
    124124
    parameters if an example violates the margin constraint, which makes
    125125
    training very efficient and may result in sparser models (i.e. with more zero
    126-
    coefficents), even when L2 penalty is used.
    126+
    coefficients), even when L2 penalty is used.
    127127

    128128
    Using ``loss="log"`` or ``loss="modified_huber"`` enables the
    129129
    ``predict_proba`` method, which gives a vector of probability estimates
    @@ -408,7 +408,7 @@ parameters, we minimize the regularized training error given by
    408408
    where :math:`L` is a loss function that measures model (mis)fit and
    409409
    :math:`R` is a regularization term (aka penalty) that penalizes model
    410410
    complexity; :math:`\alpha > 0` is a non-negative hyperparameter that controls
    411-
    the regularization stength.
    411+
    the regularization strength.
    412412

    413413
    Different choices for :math:`L` entail different classifiers or regressors:
    414414

    ‎doc/modules/svm.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -623,7 +623,7 @@ misclassified or within the margin boundary. Ideally, the value :math:`y_i
    623623
    (w^T \phi (x_i) + b)` would be :math:`\geq 1` for all samples, which
    624624
    indicates a perfect prediction. But problems are usually not always perfectly
    625625
    separable with a hyperplane, so we allow some samples to be at a distance :math:`\zeta_i` from
    626-
    their correct margin boundary. The penalty term `C` controls the strengh of
    626+
    their correct margin boundary. The penalty term `C` controls the strength of
    627627
    this penalty, and as a result, acts as an inverse regularization parameter
    628628
    (see note below).
    629629

    ‎doc/roadmap.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -51,7 +51,7 @@ external to the core library.
    5151
    (i.e. rectangular data largely invariant to column and row order;
    5252
    predicting targets with simple structure)
    5353
    * improve the ease for users to develop and publish external components
    54-
    * improve inter-operability with modern data science tools (e.g. Pandas, Dask)
    54+
    * improve interoperability with modern data science tools (e.g. Pandas, Dask)
    5555
    and infrastructures (e.g. distributed processing)
    5656

    5757
    Many of the more fine-grained goals can be found under the `API tag

    ‎doc/themes/scikit-learn-modern/static/css/theme.css

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -1237,7 +1237,7 @@ table.sk-sponsor-table td {
    12371237
    text-align: center
    12381238
    }
    12391239

    1240-
    /* pygments - highlightning */
    1240+
    /* pygments - highlighting */
    12411241

    12421242
    .highlight .hll { background-color: #ffffcc }
    12431243
    .highlight { background: #f8f8f8; }

    ‎doc/tutorial/machine_learning_map/ML_MAPS_README.txt

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -7,7 +7,7 @@ by Andreas Mueller:
    77

    88
    (https://peekaboo-vision.blogspot.de/2013/01/machine-learning-cheat-sheet-for-scikit.html)
    99

    10-
    The image is made interactive using an imagemap, and uses the jQuery Map Hilight plugin module
    10+
    The image is made interactive using an imagemap, and uses the jQuery Map Highlight plugin module
    1111
    by David Lynch (https://davidlynch.org/projects/maphilight/docs/) to highlight
    1212
    the different items on the image upon mouseover.
    1313

    ‎doc/tutorial/machine_learning_map/pyparsing.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -2836,7 +2836,7 @@ class QuotedString(Token):
    28362836
    def __init__( self, quoteChar, escChar=None, escQuote=None, multiline=False, unquoteResults=True, endQuoteChar=None, convertWhitespaceEscapes=True):
    28372837
    super(QuotedString,self).__init__()
    28382838

    2839-
    # remove white space from quote chars - wont work anyway
    2839+
    # remove white space from quote chars - won't work anyway
    28402840
    quoteChar = quoteChar.strip()
    28412841
    if not quoteChar:
    28422842
    warnings.warn("quoteChar cannot be the empty string",SyntaxWarning,stacklevel=2)

    ‎doc/whats_new/v0.16.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -54,7 +54,7 @@ Highlights
    5454

    5555
    - Out-of core learning of PCA via :class:`decomposition.IncrementalPCA`.
    5656

    57-
    - Probability callibration of classifiers using
    57+
    - Probability calibration of classifiers using
    5858
    :class:`calibration.CalibratedClassifierCV`.
    5959

    6060
    - :class:`cluster.Birch` clustering method for large-scale datasets.

    ‎doc/whats_new/v0.20.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -1286,7 +1286,7 @@ Support for Python 3.3 has been officially dropped.
    12861286
    be used for novelty detection, i.e. predict on new unseen data. Available
    12871287
    prediction methods are ``predict``, ``decision_function`` and
    12881288
    ``score_samples``. By default, ``novelty`` is set to ``False``, and only
    1289-
    the ``fit_predict`` method is avaiable.
    1289+
    the ``fit_predict`` method is available.
    12901290
    By :user:`Albert Thomas <albertcthomas>`.
    12911291

    12921292
    - |Fix| Fixed a bug in :class:`neighbors.NearestNeighbors` where fitting a

    ‎doc/whats_new/v0.21.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -1060,7 +1060,7 @@ These changes mostly affect library developers.
    10601060

    10611061
    - Add ``check_fit_idempotent`` to
    10621062
    :func:`~utils.estimator_checks.check_estimator`, which checks that
    1063-
    when `fit` is called twice with the same data, the ouput of
    1063+
    when `fit` is called twice with the same data, the output of
    10641064
    `predict`, `predict_proba`, `transform`, and `decision_function` does not
    10651065
    change. :pr:`12328` by :user:`Nicolas Hug <NicolasHug>`
    10661066

    ‎doc/whats_new/v0.23.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -341,7 +341,7 @@ Changelog
    341341
    :pr:`16006` by :user:`Rushabh Vasani <rushabh-v>`.
    342342

    343343
    - |API| The `StreamHandler` was removed from `sklearn.logger` to avoid
    344-
    double logging of messages in common cases where a hander is attached
    344+
    double logging of messages in common cases where a handler is attached
    345345
    to the root logger, and to follow the Python logging documentation
    346346
    recommendation for libraries to leave the log message handling to
    347347
    users and application code. :pr:`16451` by :user:`Christoph Deil <cdeil>`.

    ‎doc/whats_new/v0.24.rst

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -713,7 +713,7 @@ Changelog
    713713
    :user:`Joseph Willard <josephwillard>`
    714714

    715715
    - |Fix| bug in :func:`metrics.hinge_loss` where error occurs when
    716-
    ``y_true`` is missing some labels that are provided explictly in the
    716+
    ``y_true`` is missing some labels that are provided explicitly in the
    717717
    ``labels`` parameter.
    718718
    :pr:`17935` by :user:`Cary Goltermann <Ultramann>`.
    719719

    ‎examples/applications/plot_cyclical_feature_engineering.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -215,7 +215,7 @@
    215215
    # %%
    216216
    #
    217217
    # Lets evaluate our gradient boosting model with the mean absolute error of the
    218-
    # relative demand averaged accross our 5 time-based cross-validation splits:
    218+
    # relative demand averaged across our 5 time-based cross-validation splits:
    219219

    220220

    221221
    def evaluate(model, X, y, cv):

    ‎examples/calibration/plot_calibration_multiclass.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -178,7 +178,7 @@ class of an instance (red: class 1, green: class 2, blue: class 3).
    178178
    print(f" * calibrated classifier: {cal_score:.3f}")
    179179

    180180
    # %%
    181-
    # Finally we generate a grid of possibile uncalibrated probabilities over
    181+
    # Finally we generate a grid of possible uncalibrated probabilities over
    182182
    # the 2-simplex, compute the corresponding calibrated probabilities and
    183183
    # plot arrows for each. The arrows are colored according the highest
    184184
    # uncalibrated probability. This illustrates the learned calibration map:

    ‎examples/covariance/plot_mahalanobis_distances.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -70,7 +70,7 @@
    7070
    # are Gaussian distributed with mean of 0 but feature 1 has a standard
    7171
    # deviation equal to 2 and feature 2 has a standard deviation equal to 1. Next,
    7272
    # 25 samples are replaced with Gaussian outlier samples where feature 1 has
    73-
    # a standard devation equal to 1 and feature 2 has a standard deviation equal
    73+
    # a standard deviation equal to 1 and feature 2 has a standard deviation equal
    7474
    # to 7.
    7575

    7676
    import numpy as np

    ‎examples/cross_decomposition/plot_pcr_vs_pls.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -134,7 +134,7 @@
    134134
    #
    135135
    # On the other hand, the PLS regressor manages to capture the effect of the
    136136
    # direction with the lowest variance, thanks to its use of target information
    137-
    # during the transformation: it can recogize that this direction is actually
    137+
    # during the transformation: it can recognize that this direction is actually
    138138
    # the most predictive. We note that the first PLS component is negatively
    139139
    # correlated with the target, which comes from the fact that the signs of
    140140
    # eigenvectors are arbitrary.

    ‎examples/ensemble/plot_gradient_boosting_early_stopping.py

    Lines changed: 2 additions & 2 deletions
    Original file line numberDiff line numberDiff line change
    @@ -17,7 +17,7 @@
    1717
    model is trained using the training set and evaluated using the validation set.
    1818
    When each additional stage of regression tree is added, the validation set is
    1919
    used to score the model. This is continued until the scores of the model in
    20-
    the last ``n_iter_no_change`` stages do not improve by atleast `tol`. After
    20+
    the last ``n_iter_no_change`` stages do not improve by at least `tol`. After
    2121
    that the model is considered to have converged and further addition of stages
    2222
    is "stopped early".
    2323
    @@ -64,7 +64,7 @@
    6464
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2,
    6565
    random_state=0)
    6666

    67-
    # We specify that if the scores don't improve by atleast 0.01 for the last
    67+
    # We specify that if the scores don't improve by at least 0.01 for the last
    6868
    # 10 stages, stop fitting additional stages
    6969
    gbes = ensemble.GradientBoostingClassifier(n_estimators=n_estimators,
    7070
    validation_fraction=0.2,

    ‎examples/ensemble/plot_gradient_boosting_quantile.py

    Lines changed: 1 addition & 1 deletion
    Original file line numberDiff line numberDiff line change
    @@ -184,7 +184,7 @@ def highlight_min(x):
    184184
    # the fact the squared error estimator is very sensitive to large outliers
    185185
    # which can cause significant overfitting. This can be seen on the right hand
    186186
    # side of the previous plot. The conditional median estimator is biased
    187-
    # (underestimation for this asymetric noise) but is also naturally robust to
    187+
    # (underestimation for this asymmetric noise) but is also naturally robust to
    188188
    # outliers and overfits less.
    189189
    #
    190190
    # Calibration of the confidence interval

    ‎examples/inspection/plot_linear_model_coefficient_interpretation.py

    Lines changed: 2 additions & 2 deletions
    Original file line numberDiff line numberDiff line change
    @@ -354,7 +354,7 @@
    354354

    355355
    # %%
    356356
    # Two regions are populated: when the EXPERIENCE coefficient is
    357-
    # positive the AGE one is negative and viceversa.
    357+
    # positive the AGE one is negative and vice-versa.
    358358
    #
    359359
    # To go further we remove one of the 2 features and check what is the impact
    360360
    # on the model stability.
    @@ -664,7 +664,7 @@
    664664
    # It is important to keep in mind that the coefficients that have been
    665665
    # dropped may still be related to the outcome by themselves: the model
    666666
    # chose to suppress them because they bring little or no additional
    667-
    # information on top of the other features. Additionnaly, this selection
    667+
    # information on top of the other features. Additionally, this selection
    668668
    # is unstable for correlated features, and should be interpreted with
    669669
    # caution.
    670670
    #

    0 commit comments

    Comments
     (0)
    0