“Pandas oblicz IQR” Kod odpowiedzi

Pandas oblicz IQR

# Removing the outliers
def removeOutliers(data, col):
    Q3 = np.quantile(data[col], 0.75)
    Q1 = np.quantile(data[col], 0.25)
    IQR = Q3 - Q1
 
    print("IQR value for column %s is: %s" % (col, IQR))
    global outlier_free_list
    global filtered_data
 
    lower_range = Q1 - 1.5 * IQR
    upper_range = Q3 + 1.5 * IQR
    outlier_free_list = [x for x in data[col] if (
        (x > lower_range) & (x < upper_range))]
    filtered_data = data.loc[data[col].isin(outlier_free_list)]
 
 
for i in data.columns:
      if i == data.columns[0]:
      removeOutliers(data, i)
    else:
      removeOutliers(filtered_data, i)
 
  
# Assigning filtered data back to our original variable
data = filtered_data
print("Shape of data after outlier removal is: ", data.shape)