outliers_4.py

# (modified from http://www.itl.nist.gov/div898/handbook/prc/section1/prc16.htm)
def get_iqr_values(df_in, col_name):
    median = df_in[col_name].median()
    q1 = df_in[col_name].quantile(0.25) # 25th percentile / 1st quartile
    q3 = df_in[col_name].quantile(0.75) # 7th percentile / 3rd quartile
    iqr = q3-q1 #Interquartile range
    minimum  = q1-1.5*iqr # The minimum value or the |- marker in the box plot
    maximum = q3+1.5*iqr # The maximum value or the -| marker in the box plot
    return median, q1, q3, iqr, minimum, maximum

def get_iqr_text(df_in, col_name):
    median, q1, q3, iqr, minimum, maximum = get_iqr_values(df_in, col_name)
    text = f"median={median:.2f}, q1={q1:.2f}, q3={q3:.2f}, iqr={iqr:.2f}, minimum={minimum:.2f}, maximum={maximum:.2f}"
    return text

def remove_outliers(df_in, col_name):
    _, _, _, _, minimum, maximum = get_iqr_values(df_in, col_name)
    df_out = df_in.loc[(df_in[col_name] > minimum) & (df_in[col_name] < maximum)]
    return df_out

def count_outliers(df_in, col_name):
    _, _, _, _, minimum, maximum = get_iqr_values(df_in, col_name)
    df_outliers = df_in.loc[(df_in[col_name] <= minimum) | (df_in[col_name] >= maximum)]
    return df_outliers.shape[0]

def box_and_whisker(df_in, col_name):
    title = get_iqr_text(df_in, col_name)
    sns.boxplot(df_in[col_name])
    plt.title(title)
    plt.show()