۵ سوال پایتون (۱۲)

---

شرح پروژه

سلام پنج سوال پایتون دارم

مدت زمان ۱ روز

Q1. Given a dataframe with housing data, perform the following preprocessing/feature engineering steps: Remove the Id column.  Remove all non-numerical columns from the dataframe except Neighborhood.  If a column has >= 60% NaN values, remove the columns from the dataset.  Otherwise, for numerical columns, impute those columns with the statistical median of the particular column.  Apply a Min Max Scaler to all the numerical features in the dataframe except Neighborhood.  Use the DictVectorizer to encode Neighborhood into one-hot feature representations and name these additional columns as f0,f1,f2..etc. to replace the Neighborhood column.  Return the final dataframe.

Q2. Given a dataframe with multiple features f0, f1, ..., apply KMeans for the given values of k and return the value of k with the least inertia of the clustering and this least intertia value.

Q3. Given a dataframe with housing data to predict the SalePrice value, return the best features (NOT SalePrice) using Lasso feature selection method. Use Lasso and SelectFromModel to achieve this. Input : df - dataframe and alpha parameter for Lasso Return : Set (set()) of columns with the feature names selected by Lasso

Q4. Given a dataframe with housing data to predict the SalePrice value, return the average value of the validation mean squared error and alpha value of the best Ridge model fit in a  k -fold Cross Validation setting based on a given set of alpha values (of the Ridge model).Q4. Given a dataframe with housing data to predict the SalePrice value, return the average value of the validation mean squared error and alpha value of the best Ridge model fit in a  k -fold Cross Validation setting based on a given set of alpha values (of the Ridge model).

Q5. Given a dataframe about results from a drug trial with two drugs (drug_type 0 and 1) that aim to improve longevity (life_expectancy), use bootstrap estimation to find the confidence interval of difference in mean value of life_expectancy of each of these two drugs. Use the pandas Dataframe.sample function to sample with replacement, with the bootstrap sample size as the length of the source dataframe and random state passed in the function argument. Use drug_type 0 - drug_type 1order for the difference. Arguments  df : a dataframe that includes observations of the two sample classes variable : the column name of the column that includes observations of the variable of interest class_name : the column name of the column that includes group assignment (This column should contain two different group names/values) num_repetitions: number of times you want the bootstrapping to repeat. Default is 1000. alpha : likelihood that the true population parameter lies outside the confidence interval. Default is 0.05. random_state : enable users to set their own random_state for the sampling, default is 123. Return  The lower limit and upper limit of the confidence interval for the difference in mean life_expectancy between the two drug types corresponding to the given alpha.Q5. Given a dataframe about results from a drug trial with two drugs (drug_type 0 and 1) that aim to improve longevity (life_expectancy), use bootstrap estimation to find the confidence interval of difference in mean value of life_expectancy of each of these two drugs. Use the pandas Dataframe.sample function to sample with replacement, with the bootstrap sample size as the length of the source dataframe and random state passed in the function argument. Use drug_type 0 - drug_type 1order for the difference. Arguments  df : a dataframe that includes observations of the two sample classes variable : the column name of the column that includes observations of the variable of interest class_name : the column name of the column that includes group assignment (This column should contain two different group names/values) num_repetitions: number of times you want the bootstrapping to repeat. Default is 1000. alpha : likelihood that the true population parameter lies outside the confidence interval. Default is 0.05. random_state : enable users to set their own random_state for the sampling, default is 123. Return  The lower limit and upper limit of the confidence interval for the difference in mean life_expectancy between the two drug types corresponding to the given alpha.

DO NOT IMPORT ANYTHING OTHER THAN SUB-PACKAGES OF THESE WHEN NECESSARY.

import numpy as np import pandas as pd  from sklearn import preprocessing  from sklearn import feature_extraction  from sklearn import feature_selection   from sklearn import linear_model from sklearn import cluster  from sklearn import model_selection from sklearn import metrics

housing_raw = pd.read_csv('https://raw.githubusercontent.com/ift-6758/files/main/housing_raw.csv')  cluster_data = pd.read_csv('https://raw.githubusercontent.com/ift-6758/files/main/cluster-data.csv')  housing_processed = pd.read_csv('https://raw.githubusercontent.com/ift-6758/files/main/housing_processed.csv')  drug_trials = pd.read_csv('https://raw.githubusercontent.com/ift-6758/files/main/drug_trials.csv')

مهارت ها و تخصص های مورد نیاز

---

---