How to use iqr_outliers method in pytest-benchmark

Best Python code snippet using pytest-benchmark

outliers.py

Source:outliers.py

...14        self.replace = replace15    # TODO: Tal vez implementar eliminaciÃ³n de outliers 2D para latitud y longitud16    # TODO: En los inputers tengo replacement por moda, mean, etc podrÃa disponibilizar mÃ©todos desde acÃ¡17    RIDICULOUS_PRICE_FOR_PROPERTY = 4000018    def iqr_outliers(19        self, data: pd.DataFrame, field: str, strategy="mode", margin: float = 1.520    ) -> pd.DataFrame:21        """MÃ©todo de remociÃ³n de outliers usando IQR"""22        # SerÃ¡ la primera barrera de eliminaciÃ³n, solo tenemos campos positivos23        if not data[data[field].isnull()].empty:24            raise RuntimeError(25                f"Para aplicar IQR asegurate que el campo '{field}' no contenga NaN"26            )27        Q1 = data[field].quantile(0.25)28        Q3 = data[field].quantile(0.75)29        IQR = Q3 - Q130        mask = ~(31            (data[field] < (Q1 - margin * IQR)) | (data[field] > (Q3 + margin * IQR))32        )33        if self.replace:34            if strategy == "mode":35                replace = data[field].mode()[0]36            elif strategy == "mean":37                replace = data[field].mean()38            else:39                replace = data[field].median()40            data.loc[mask, field] = data.loc[mask, field].fillna(replace)41            return data42        return data[mask]43    def iqr_outliers_prices_per_comuna(self, data: pd.DataFrame) -> pd.DataFrame:44        """Filtra outliers por cada comuna 'l3'45        The idea behind removing localized outliers by 'l3' is the socioeconomic46        level, the prices of communes more exclusive areas,47        such as Puerto Madero or Recoleta, would be affected when48        trimming with global IQR. A possible better one is maybe49        do not cut by 'l3', but by larger areas, like south zone, central zone50        and north zone (It is a TODO!)51        """52        data = data[data.price > self.RIDICULOUS_PRICE_FOR_PROPERTY]53        groups = [54            self.iqr_outliers(group, "price")55            for _, group in data.copy().groupby("Comuna")56        ]57        return pd.concat(groups)58    def fit(self, X: pd.DataFrame, y=None) -> FieldOutliers:59        return self60    def transform(self, X: pd.DataFrame) -> pd.DataFrame:61        X = X.copy()62        X = self.iqr_outliers(X, "rooms", margin=3.2)63        X = self.iqr_outliers(X, "bedrooms", margin=3.2)64        X = self.iqr_outliers(X, "bathrooms", margin=3.2)65        X = self.iqr_outliers(X, "surface_total", strategy="mean")...

anomaly_detection_methodes.py

Source:anomaly_detection_methodes.py

1from scipy import stats2import numpy as np3import pandas as pd45def anomaly_detector(data, config):6    if config['time_series']:7        # TODO8        return None9    else:10        id_column=data['id']11        data=data.drop(columns=['id'])12        IQR_outliers=iqr_outlier_detector(data)13        zscore_outliers=zscore_outlier_detector(data)14        columns=['id']15        for i in range(len(zscore_outliers.columns)):16            columns.append(IQR_outliers.columns[i])17            columns.append(zscore_outliers.columns[i])18        result=pd.DataFrame(columns=columns)19        print(columns)20        result['id']=id_column.values21        result[IQR_outliers.columns]=IQR_outliers.values22        result[zscore_outliers.columns]=zscore_outliers.values2324    return result252627def iqr_outlier_detector(data):28    columns=data.columns29    columns=list(map(lambda x: 'IQR_method_'+x,columns))30    # calculate Q1 and Q331    Q1 = data.quantile(0.25)32    Q3 = data.quantile(0.75)3334    # calculate the IQR35    IQR = Q3 - Q13637    # filter the dataset with the IQR38    IQR_outliers = (data < (Q1 - 1.5 * IQR)) |(data > (Q3 + 1.5 * IQR))39    IQR_outliers.columns=columns40    return IQR_outliers4142def zscore_outlier_detector(data):43    columns=data.columns44    columns=list(map(lambda x: 'zscore_method_'+x,columns))45    data=(data-data.mean(axis=0))/data.std(axis=0)46    zscore_outliers= data >= 347    zscore_outliers.columns=columns
...

outlier_test_plot.py

Source:outlier_test_plot.py

...6data = get_image(3,9).get_data()7std = vol_std(data)8# RMS outliers plot 9rms = vol_rms_diff(data)10outliers_rms = iqr_outliers(rms)[0]11ranges = iqr_outliers(rms)[1]12outliers_values = []13for i in outliers_rms:14	outliers_values.append(rms[i])15plt.figure()16plt.plot(rms)17plt.plot(outliers_rms,outliers_values,"ro", marker = "o")18plt.axhline(y = ranges[0], linestyle="--", color = "g")19plt.axhline(y = ranges[1], linestyle="--", color = "g")20plt.savefig('outliers RMS sub9 run3')21### subject 1 run 1 22data = get_image(1,1).get_data()23std = vol_std(data)24rms = vol_rms_diff(data)25outliers_rms = iqr_outliers(rms)[0]26ranges = iqr_outliers(rms)[1]27outliers_values = []28for i in outliers_rms:29	outliers_values.append(rms[i])30plt.figure()31plt.plot(rms)32plt.plot(outliers_rms,outliers_values,"ro", marker = "o")33plt.axhline(y = ranges[0], linestyle="--", color = "g")34plt.axhline(y = ranges[1], linestyle="--", color = "g")...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.