Please complete the following exercise using Python or R and submit the full code along withyour results.1. Download the csv file with the data...

Please complete the following exercise using Python or R and submit the full code along withyour results.1. Download the csv file with the data fromhttps://github.com/adebayoj/fairml/blob/master/doc/example_notebooks/propublica_data_for_fairml.csv directly into your code. Any lines with missing data should be deleted.The data contains information concerning attributes of a group of convicts in the United States,including racial identification, age group, sex, number of priors, and whether or not their offensewas a misdemeanor. An ML model was used to predict whether each convict would reoffend inthe short run. The binary variable score_report represents the result of the model’s prediction andanother binary variable, Two_yr_Recidivism, indicates whether the convict actually reoffended inthe subsequent two years.2. Assume that the convicts that do not identify with any of the mentioned racial categories(African American, Asian, Hispanic, Native American, Other) identify as white. Usingthis criterion only, add a column to the data set indicating which convicts identify aswhite.3. The convicts are divided into 4 categories according to their number of priors.Category Number of PriorsA 0B 1-3C 4-10D More than 10Add a column to the data set indicating the category of each convict.4. Run regression models to determine the influence of the variables Number_of_Priors,Age_Below_TwentyFive, African_American, Hispanic, Asian, Native_American,Female, and Misdeamenor on the variables Two_yr_Recidivism and score_factor.Conclude on your results.5. Regressions allow one to measure a variables’ dependence on particular attributes.However, indirect effects may not be accounted for when the attributes used in the modelare co-linear. Comment on whether this is the case in your regression models and why.6. What experiments or analysis would you conduct in order to determine whether there isracial, gender, and/or age bias in the way that the ML model is predicting recidivism?