h$,fV ddlZddlmZmZddlZddlmZmZm Z m Z m Z m Z ddl mZddlmZmZddlmZmZmZmZmZddlmZmZdd lmZdd lmZmZm Z d d l!m"Z"ejFd Z$d Z%Gddeee e eZ&y)N)IntegralReal) BaseEstimatorMetaEstimatorMixinMultiOutputMixinRegressorMixin _fit_contextclone)ConvergenceWarning)check_consistent_lengthcheck_random_state) HasMethodsIntervalOptions RealNotInt StrOptions)_raise_for_unsupported_routing_RoutingNotSupportedMixin)sample_without_replacement)_check_sample_weightcheck_is_fittedhas_fit_parameter)LinearRegressionc <|t|z }ttd|z }ttd||zz }|dk(ry|dk(r tdSttt j t j |t j |z S)aDetermine number trials such that at least one outlier-free subset is sampled for the given inlier/outlier ratio. Parameters ---------- n_inliers : int Number of inliers in the data. n_samples : int Total number of samples in the data. min_samples : int Minimum number of samples chosen randomly from original data. probability : float Probability (confidence) that one outlier-free sample is generated. Returns ------- trials : int Number of trials. rrinf)floatmax_EPSILONabsnpceillog) n_inliers n_samples min_samples probability inlier_rationomdenoms `. Parameters ---------- estimator : object, default=None Base estimator object which implements the following methods: * `fit(X, y)`: Fit model to given training data and target values. * `score(X, y)`: Returns the mean accuracy on the given test data, which is used for the stop criterion defined by `stop_score`. Additionally, the score is used to decide which of two equally large consensus sets is chosen as the better one. * `predict(X)`: Returns predicted values using the linear model, which is used to compute residual error using loss function. If `estimator` is None, then :class:`~sklearn.linear_model.LinearRegression` is used for target values of dtype float. Note that the current implementation only supports regression estimators. min_samples : int (>= 1) or float ([0, 1]), default=None Minimum number of samples chosen randomly from original data. Treated as an absolute number of samples for `min_samples >= 1`, treated as a relative number `ceil(min_samples * X.shape[0])` for `min_samples < 1`. This is typically chosen as the minimal number of samples necessary to estimate the given `estimator`. By default a :class:`~sklearn.linear_model.LinearRegression` estimator is assumed and `min_samples` is chosen as ``X.shape[1] + 1``. This parameter is highly dependent upon the model, so if a `estimator` other than :class:`~sklearn.linear_model.LinearRegression` is used, the user must provide a value. residual_threshold : float, default=None Maximum residual for a data sample to be classified as an inlier. By default the threshold is chosen as the MAD (median absolute deviation) of the target values `y`. Points whose residuals are strictly equal to the threshold are considered as inliers. is_data_valid : callable, default=None This function is called with the randomly selected data before the model is fitted to it: `is_data_valid(X, y)`. If its return value is False the current randomly chosen sub-sample is skipped. is_model_valid : callable, default=None This function is called with the estimated model and the randomly selected data: `is_model_valid(model, X, y)`. If its return value is False the current randomly chosen sub-sample is skipped. Rejecting samples with this function is computationally costlier than with `is_data_valid`. `is_model_valid` should therefore only be used if the estimated model is needed for making the rejection decision. max_trials : int, default=100 Maximum number of iterations for random sample selection. max_skips : int, default=np.inf Maximum number of iterations that can be skipped due to finding zero inliers or invalid data defined by ``is_data_valid`` or invalid models defined by ``is_model_valid``. .. versionadded:: 0.19 stop_n_inliers : int, default=np.inf Stop iteration if at least this number of inliers are found. stop_score : float, default=np.inf Stop iteration if score is greater equal than this threshold. stop_probability : float in range [0, 1], default=0.99 RANSAC iteration stops if at least one outlier-free set of the training data is sampled in RANSAC. This requires to generate at least N samples (iterations):: N >= log(1 - probability) / log(1 - e**m) where the probability (confidence) is typically set to high value such as 0.99 (the default) and e is the current fraction of inliers w.r.t. the total number of samples. loss : str, callable, default='absolute_error' String inputs, 'absolute_error' and 'squared_error' are supported which find the absolute error and squared error per sample respectively. If ``loss`` is a callable, then it should be a function that takes two arrays as inputs, the true and predicted value and returns a 1-D array with the i-th value of the array corresponding to the loss on ``X[i]``. If the loss on a sample is greater than the ``residual_threshold``, then this sample is classified as an outlier. .. versionadded:: 0.18 random_state : int, RandomState instance, default=None The generator used to initialize the centers. Pass an int for reproducible output across multiple function calls. See :term:`Glossary `. Attributes ---------- estimator_ : object Best fitted model (copy of the `estimator` object). n_trials_ : int Number of random selection trials until one of the stop criteria is met. It is always ``<= max_trials``. inlier_mask_ : bool array of shape [n_samples] Boolean mask of inliers classified as ``True``. n_skips_no_inliers_ : int Number of iterations skipped due to finding zero inliers. .. versionadded:: 0.19 n_skips_invalid_data_ : int Number of iterations skipped due to invalid data defined by ``is_data_valid``. .. versionadded:: 0.19 n_skips_invalid_model_ : int Number of iterations skipped due to an invalid model defined by ``is_model_valid``. .. versionadded:: 0.19 n_features_in_ : int Number of features seen during :term:`fit`. .. versionadded:: 0.24 feature_names_in_ : ndarray of shape (`n_features_in_`,) Names of features seen during :term:`fit`. Defined only when `X` has feature names that are all strings. .. versionadded:: 1.0 See Also -------- HuberRegressor : Linear regression model that is robust to outliers. TheilSenRegressor : Theil-Sen Estimator robust multivariate regression model. SGDRegressor : Fitted by minimizing a regularized empirical loss with SGD. References ---------- .. [1] https://en.wikipedia.org/wiki/RANSAC .. [2] https://www.sri.com/wp-content/uploads/2021/12/ransac-publication.pdf .. [3] http://www.bmva.org/bmvc/2009/Papers/Paper355/Paper355.pdf Examples -------- >>> from sklearn.linear_model import RANSACRegressor >>> from sklearn.datasets import make_regression >>> X, y = make_regression( ... n_samples=200, n_features=2, noise=4.0, random_state=0) >>> reg = RANSACRegressor(random_state=0).fit(X, y) >>> reg.score(X, y) 0.9885... >>> reg.predict(X[:1,]) array([-31.9417...]) )fitscorepredictNrleft)closedrbothabsolute_error squared_error random_state) estimatorr'residual_threshold is_data_validis_model_valid max_trials max_skipsstop_n_inliers stop_scorestop_probabilitylossr9_parameter_constraintsdgGz?) r'r;r<r=r>r?r@rArBrCr9c ||_||_||_||_||_||_||_||_| |_| |_ | |_ | |_ yN) r:r'r;r<r=r>r?r@rArBr9rC) selfr:r'r;r<r=r>r?r@rArBrCr9s r,__init__zRANSACRegressor.__init__s_ #&"4*,$",$ 0( r.F)prefer_skip_nested_validationc < t|d|tdd}td}|j||||f\}}t|||jt |j}n t }|j.t|t s td |jd d z}ncd |jcxkrd kr3nn0tj|j|jd z}n|jd k\r |j}|jd kDrtd |jd z|j?tjtj|tj|z }n |j}|j d k(r|j"d k(rd} nKd} nG|j dk(r|j"d k(rd} n%d} n!t%|j r |j } t'|j(} |j+| t-|d} t/|j0} || std| z| t3||}d } tj4 }d}d}d}d}d |_d |_d |_|jd }tj<|}d |_|j@}|j>|kr|xj>d z c_|j6|j8z|j:z|jBkDrntE||| }||}||}|jF(|jG||s|xj8d z c_||jI||n|jI|||||jJ*|jK|||s|xj:d z c_|jM|} ||}||k}tjN|}|| kr|xj6d z c_X||}||}||}|jQ||} || k(r| |kr|} | }|}|}|}|}tS|tU| |||jV}| |jXk\s||jZk\rn|j>|kr|I|j6|j8z|j:z|jBkDr tdtd|j6|j8z|j:z|jBkDrt]j^dt`||jI||n|jI||||||_1||_2|S#t$rYkwxYw)aFit estimator using RANSAC algorithm. Parameters ---------- X : {array-like, sparse matrix} of shape (n_samples, n_features) Training data. y : array-like of shape (n_samples,) or (n_samples, n_targets) Target values. sample_weight : array-like of shape (n_samples,), default=None Individual weights for each sample raises error if sample_weight is passed and estimator fit method does not support it. .. versionadded:: 0.18 Returns ------- self : object Fitted `RANSACRegressor` estimator. Raises ------ ValueError If no valid consensus set could be found. This occurs if `is_data_valid` and `is_model_valid` return False for all `max_trials` randomly chosen sub-samples. r1) sample_weightcsrF) accept_sparseforce_all_finite) ensure_2d)validate_separatelyNzR`min_samples` needs to be explicitly set when estimator is not a LinearRegression.rrzG`min_samples` may not be larger than number of samples: n_samples = %d.r7c2tj||z SrG)r"r!y_truey_preds r,z%RANSACRegressor.fit..|srvvfvo7Nr.c\tjtj||z dS)Nraxis)r"sumr!rSs r,rVz%RANSACRegressor.fit..~s!rvvFF6F?+!8r.r8c||z dzS)NrrSs r,rVz%RANSACRegressor.fit..sA7Mr.c<tj||z dzdS)NrrrX)r"rZrSs r,rVz%RANSACRegressor.fit..srvvf_*8r.)r9rLz\%s does not support sample_weight. Samples weights are only used for the calibration itself.zRANSAC skipped more iterations than `max_skips` without finding a valid consensus set. Iterations were skipped because each randomly chosen sub-sample failed the passing criteria. See estimator attributes for diagnostics (n_skips*).zRANSAC could not find a valid consensus set. All `max_trials` iterations were skipped because each randomly chosen sub-sample failed the passing criteria. See estimator attributes for diagnostics (n_skips*).zRANSAC found a valid consensus set but exited early due to skipping more iterations than `max_skips`. See estimator attributes for diagnostics (n_skips*).)3rdict_validate_datar r:r rr' isinstance ValueErrorshaper"r#r;medianr!rCndimcallablerr9 set_paramsrtype__name__rrn_skips_no_inliers_n_skips_invalid_data_n_skips_invalid_model_arange n_trials_r>r?rr<r1r=r3rZr2minr-rBr@rAwarningswarnr estimator_ inlier_mask_)!rHXyrLcheck_X_paramscheck_y_paramsr:r'r; loss_functionr9estimator_fit_has_sample_weightestimator_namen_inliers_best score_bestinlier_mask_best X_inlier_best y_inlier_bestinlier_best_idxs_subsetr& sample_idxsr> subset_idxsX_subsety_subsetrUresiduals_subsetinlier_mask_subsetn_inliers_subsetinlier_idxs_subsetX_inlier_subsety_inlier_subset score_subsets! r,r1zRANSACRegressor.fit1sD 'tU-PEEJ."" q~~&F# 1 1% >> %dnn-I(*I    #i)9: 1''!*q.K !! %A %''$"2"2QWWQZ"?@K    "**K  #.12=   " " *!#266!biil2B+C!D !%!8!8  99( (vv{ N ! YY/ )vv{ M ! dii  IIM)$*;*;<    l ;+1lZ6O.N%J1 +M+M&8 ##"I{Drso ",?6WW# 2::a=