you can use python or matlab1 CAP 5625: Programming Assignment 2 Due on Canvas by Wednesday,...

Question

you can use python or matlab1    CAP 5625: Programming Assignment 2    Due on Canvas by Wednesday, November 3, 2021 at 11:59pm    Preliminary instructions  You may consult with other students currently taking CAP 5625 at FAU on this programming  assignment. If you do consult with others, then you must indicate this by providing their names  with your submitted assignment. However, all analyses must be performed independently, all  source code must be written independently, and all students must turn in their own  independent assignment.    Though it should be unnecessary to state in a graduate class, I am reminding you  that you may not turn in code (partial or complete) that is written or inspired by  others, including code from other students, websites, past code that I release  from prior assignments in this class or from past semesters in other classes I  teach, or any other source that would constitute an academic integrity violation.  All instances of academic integrity violations will receive a zero on the  assignment and will be referred to the Department Chair and College Dean for  further administrative action.    You may choose to use whatever programming language you want. However, you must provide  clear instructions on how to compile and/or run your source code. I recommend using a  modern language, such as Python, R, or Matlab as learning these languages can help you if you  were to enter the machine learning or artificial intelligence field in the future.    All analyses performed and algorithms run must be written from scratch. That is, you may not  use a library that can perform coordinate descent, cross validation, elastic net, least squares  regression, optimization, etc. to successfully complete this programing assignment (though you  may reuse your relevant code from Programming Assignment 1). The goal of this assignment is  not to learn how to use particular libraries of a language, but it is to instead understand how  key methods in statistical machine learning are implemented. With that stated, I will provide  10% extra credit if you additionally implement the assignment using built-in statistical or  machine learning libraries (see Deliverable 6 at end of the document).    Brief overview of assignment  In this assignment you will still be analyzing the same credit card data from ? = 400 training  observations that you examined in Programming Assignment 1. The goal is to fit a model that  can predict credit balance based on ? = 9 features describing an individual, which include an  individual’s income, credit limit, credit rating, number of credit cards, age, education level,  gender, student status, and marriage status. Specifically, you will perform a penalized  (regularized) least squares fit of a linear model using elastic net, with the model parameters  obtained by coordinate descent. Elastic net will permit you to provide simultaneous parameter  shrinkage (tuning parameter ? ≥ 0) and feature selection (tuning parameter ? ∈ [0,1]). The  2    two tuning parameters ? and ? will be chosen using five-fold cross validation, and the best-fit  model parameters will be inferred on the training dataset conditional on an optimal pair of  tuning parameters.    Data  Data for these observations are given in Credit_N400_p9.csv, with individuals labeled on  each row (rows 2 through 401), and input features and response given on the columns (with  the first row representing a header for each column). There are six quantitative features, given  by columns labeled “Income”, “limit”, “Rating”, “Cards”, “Age”, and “Education”, and three  qualitative features with two levels labeled “Gender”, “Student”, and “Married”.    Detailed description of the task  Recall that the task of performing an elastic net fit to training data  {(?1, ?1), (?2, ?2), … , (?? , ??)} is to minimize the cost function  ?(?, ?, ?) =  ∑(?? −∑????? ? ?=1 ) 2 ? ?=1 + ?(?∑?? 2 ? ?=1 + (1 − ?)∑|??| ? ?=1 )  where ?? is a centered response and where the input ? features are standardized (i.e., centered  and divided by their standard deviation). Note that we cannot use gradient descent to minimize  this cost function, as the component ∑ |??| ? ?=1  of the penalty is not differentiable. Instead, we  use coordinate descent, where we update each parameter ?, ? = 1,2, … , ?, in turn, keeping all  other parameters constant, and using sub-gradient rather than gradient calculations. To  implement this algorithm, depending on whether your chosen language can quickly compute  vectorized operations, you may implement coordinate descent using either Algorithm 1 or  Algorithm 2 below (choose whichever you are more comfortable implementing). Note that in  languages like R, Python, or Matlab, Algorithm 2 (which would be implemented by several  nested loops) may be much slower than Algorithm 1. Also note that if you are implementing  Algorithm 1 using Python, use numpy arrays instead of Pandas data frames for computational  speed. For this assignment, assume that we will reach the minimum of the cost function within  a fixed number of steps, with the number of iterations being 1000.       3    Algorithm 1 (vectorized):  Step 1. Fix tuning parameters ? and ?  Step 2. Generate ?-dimensional centered response vector ? and ? × ? standardized  (centered and scaled to have unit standard deviation) design matrix ?  Step 3. Precompute ??, ? = 1,2, … , ?, as  ?? =∑??? 2 ? ?=1   Step 4. Randomly initialize the parameter vector ? = [?1, ?2, … , ??]  Step 5. For each ?, ? = 1,2, … , ?:   compute  ?? = x? ?(? − ?? + x???)  and set  ?? = sign(??) (|??| − ?(1 − ?) 2 ) + ?? + ??   Step 6. Repeat Step 5 for 1000 iterations or until convergence (vector ? does not change)  Step 7. Set the last updated parameter vector as �̂� = [�̂�1, �̂�2, … , �̂�?]  4    Algorithm 2 (non-vectorized):  Step 1. Fix tuning parameters ? and ?  Step 2. Generate ?-dimensional centered response vector ? and ? × ? standardized  (centered and scaled to have unit standard deviation) design matrix ?   Step 3. Precompute ??, ? = 1,2, … , ?, as  ?? =∑??? 2 ? ?=1   Step 4. Randomly initialize the parameter vector ? = [?1, ?2, … , ??]  Step 5. For each ?, ? = 1,2, … , ?:   compute  ?? =∑??? (     ?? −∑????? ? ?=1 ?≠? )     ? ?=1   and set  ?? = sign(??) (|??| − ?(1 − ?) 2 ) + ?? + ??   Step 6. Repeat Step 5 for 1000 iterations or until convergence (vector ? does not change)  Step 7. Set the last updated parameter vector as �̂� = [�̂�1, �̂�2, … , �̂�?]    Note that we define  sign(?) = { −1 if ?

Sandeep Kumar · Accepted Answer

ass2.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "id": "ZfXIWZ57NkON"
   },
   "outputs": [],
   "source": [
    "import numpy as np
",
    "import matplotlib.pyplot as plt
",
    "import os
",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "id": "nMdVQfNePLQJ"
   },
   "outputs": [],
   "source": [
    "class Elastic:
",
    "    def __init__(self, preds, responses, lambdas, alphas, nf = 5, iters = 1000, to_verb = False):
",
    "
",
    "        self.x, self.y = self._shuffle_data(preds, responses)
",
    "        self.lambdas = lambdas 
",
    "        self.alphas = alphas
",
    "        self.nf = nf 
",
    "        self.iters = iters
",
    "        self.to_verb = to_verb
",
    "        self._initialize()
",
    "
",
    "
",
    "    def _initialize(self):
",
    "        '''
",
    "        This function initializes certain values needed for training and checks that certain
",
    "        arguments have appropriate values. 
",
    "        '''
",
    "
",
    "        # get number of samples and number of features
",
    "        self.n_samples = self.x.shape[0]
",
    "        self.n_preds = self.x.shape[1]
",
    "    
",
    "        # matrix to store the cross validation results 
",
    "        self.cv_vals = np.zeros([self.nf, len(self.lambdas), len(self.alphas)])
",
    "
",
    "        # determine the number of validation samples and their inds based on nf 
",
    "        self.n_val_samples = self.n_samples // self.nf 
",
    "        self.val_inds = list(range(0, self.n_samples, self.n_val_samples))
",
    "        
",
    "        if self.to_verb:
",
    "            print('[INFO] Using {} training and {} validation samples for each CV fold.'.format(
",
    "                self.n_samples - self.n_val_samples, self.n_val_samples)
",
    "            )
",
    "
",
    "        # create a tensor to store the Betas 
",
    "        self.B_trained = np.zeros([self.nf, len(self.lambdas), len(self.alphas), self.n_preds])
",
    "
",
    "    def _standardize(self, x, mean_vec, std_vec):
",
    "
",
    "
",
    "        return (x - mean_vec) / std_vec 
",
    "
",
    "    def _center_responses(self, y, mean):
",
    "
",
    "
",
    "        return y - mean
",
    "
",
    "    def _shuffle_data(self, preds, responses):
",
    "
",
    "
",
    "        data = np.concatenate((preds, responses[:, None]), 1)
",
    "        np.random.shuffle(data)
",
    "        return data[:, :-1], data[:, -1]
",
    "
",
    "    def _initialize_B(self):
",
    "
",
    "
",
    "        return np.random.uniform(low = -1, high = 1, size = (self.n_preds, 1))
",
    "
",
    "    def predict(self, x):
",
    "
",
    "
",
    "        assert(self.mean_vec is not None and self.std_vec is not None), \
",
    "            'Model must be trained before predicting.'
",
    "
",
    "        x = self._standardize(x, self.mean_vec, self.std_vec)
",
    "        return np.matmul(x, self.B)
",
    "
",
    "    def _get_folds(self, val_ind):
",
    " 
",
    "
",
    "        x_val = self.x[val_ind:val_ind + self.n_val_samples]
",
    "        x_train = np.delete(self.x, np.arange(val_ind, val_ind + self.n_val_samples), axis = 0)
",
    "        y_val = self.y[val_ind:val_ind + self.n_val_samples]
",
    "        y_train = np.delete(self.y, np.arange(val_ind, val_ind + self.n_val_samples), axis = 0)
",
    "        return x_train, x_val, y_train, y_val
",
    "
",
    "    def _update(self, x, y, B, ss_cols, lambda_, alpha):
",
    "
",
    "        
",
    "        for k in range(self.n_preds):
",
    "            # get rss without the effect of coefficient k
",
    "            rss_k = y - np.matmul(x, B) + (x[:, k] * B[k])[:, None]
",
    "            
",
    "            # a_k is part of the derivative of the rss term in the loss function
",
    "            a_k = np.matmul(x[:, k].transpose(), rss_k)[0]
",
    "            
",
    "            # update B_k
",
    "            B_k = np.absolute(a_k) - lambda_ * (1 - alpha) / 2
",
    "            B_k = B_k if B_k >= 0 else 0
",
    "            B[k, 0] = np.sign(a_k) * B_k / (ss_cols[k] + lambda_ * alpha)
",
    "        
",
    "        return B
",
    "
",
    "    def score(self, x, y, B):
",
    "
",
    "
",
    "        y_hat = np.matmul(x, B)
",
    "        mse = np.mean((y - y_hat) ** 2)
",
    "        return mse
",
    "    
",
    "    def _compute_ss_cols(self, x):
",
    "        '''
",
    "        Computes the sum of squares for each column needed in the update.
",
    "        Only need to do this once per fold at the beginning. 
",
    "        
",
    "        Args:
",
    "            x (np.ndarray): The N x M design matrix of the current fold. 
",
    "            
",
    "        Returns:
",
    "            ss_cols (np.ndarray): The M-dim vector representing the sum
",
    "                of squares of the cols of x.
",
    "        '''
",
    "        return np.sum(x ** 2, 0)
",
    "    
",
    "    def _find_best_tuning_params(self):
",
    "        '''
",
    "        Finds the value of lambda and alpha with minimum corresponding CV score and
",
    "        saves them as class attributes. 
",
    "        '''
",
    "        cv_mean = np.mean(self.cv_vals, 0)
",
    "        best_lambda_ind, best_alpha_ind = np.where(cv_mean == np.amin(cv_mean))
",
    "        self.best_lambda = self.lambdas[best_lambda_ind[0]]
",
    "        self.best_alpha = self.alphas[best_alpha_ind[0]]
",
    "
",
    "    def fit(self):
",
    "        '''
",
    "        Implements the cross validation and retrains the model with the optimal lambda 
",
    "        and alpha on the entire dataset.
",
    "        '''
",
    "
",
    "        for i_lambda, lambda_ in enumerate(self.lambdas): # loop through lambdas
",
    "            for i_alpha, alpha in enumerate(self.alphas): # loop through alphas
",
    "                for i_fold, val_ind in zip(range(self.nf), self.val_inds): # loop through folds
",
    "                    # get the folds
",
    "                    x_train, x_val, y_train, y_val = self._get_folds(val_ind)
",
    "
",
    "                    # standardize x and center y
",
    "                    mean_vec, std_vec = np.mean(x_train, 0), np.std(x_train, 0)
",
    "                    mean_response = np.mean(y_train)
",
    "                    x_train = self._standardize(x_train, mean_vec, std_vec)
",
    "                    x_val = self._standardize(x_val, mean_vec, std_vec)
",
    "                    y_train = self._center_responses(y_train, mean_response)[:, None]
",
    "                    y_val = self._center_responses(y_val, mean_response)[:, None]
",
    "                    
",
    "                    # compute b_k given this fold -- don't need to compute every update
",
    "                    ss_cols = self._compute_ss_cols(x_train)
",
    "
",
    "                    # initialize Beta for this lambda and fold
",
    "                    B = self._initialize_B()
",
    "
",
    "                    for iter in range(self.iters):
",
    "                        B = self._update(x_train, y_train, B, ss_cols, lambda_, alpha)
",
    "                        
",
    "                    # score this model 
",
    "                    score = self.score(x_val, y_val, B)
",
    "
",
    "                    # store the score with the tuning param combinations
",
    "                    self.cv_vals[i_fold, i_lambda, i_alpha] = score
",
    "
",
    "                    # store the coefficient vector
",
    "                    self.B_trained[i_fold, i_lambda, i_alpha] = B[:, 0]
",
    "                    
",
    "                # if to_verb flag, then print out the mean CV MSE for the combo of lambda and alpha
",
    "                if self.to_verb:
",
    "                    print('lambda:{}; alpha:{}; CV MSE:{}'.format(
",
    "                        lambda_, alpha, np.mean(self.cv_vals[:, i_lambda, i_alpha]))
",
    "                    )
",
    "
",
    "
",
    "        ############# Retrain on entire dataset with optimal lambda and alpha #############
",
    "                    
",
    "        # find the best lambda and alpha
",
    "        self._find_best_tuning_params()
",
    "        
",
    "        # standardize features of x and center responses 
",
    "        self.mean_vec, self.std_vec = np.mean(self.x, 0), np.std(self.x, 0)
",
    "        x = self._standardize(self.x, self.mean_vec, self.std_vec)
",
    "        y = self._center_responses(self.y, np.mean(self.y))[:, None]
",
    "        
",
    "        # compute the sum of squares for each feature on the entire dataset
",
    "        ss_cols = self._compute_ss_cols(x)
",
    "        
",
    "        # initialize coefficients
",
    "        self.B = self._initialize_B()
",
    "        
",
    "        # perform updates 
",
    "        for iter in range(self.iters):
",
    "            self.B = self._update(x, y, self.B, ss_cols, self.best_lambda, self.best_alpha)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 419
    },
    "id": "2yKcu-xHOwnO",
    "outputId": "6b6549d7-3105-4600-c9a1-1bb42d9f335c"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "
",
       "
",
       "    .dataframe tbody tr th:only-of-type {
",
       "        vertical-align: middle;
",
       "    }
",
       "
",
       "    .dataframe tbody tr th {
",
       "        vertical-align: top;
",
       "    }
",
       "
",
       "    .dataframe thead th {
",
       "        text-align: right;
",
       "    }
",
       "
",
       "
",
       "  
",
       "    
",
       "      
",
       "      Income
",
       "      Limit
",
       "      Rating
",
       "      Cards
",
       "      Age
",
       "      Education
",
       "      Gender
",
       "      Student
",
       "      Married
",
       "      Balance
",
       "    
",
       "  
",
       "  
",
       "    
",
       "      0
",
       "      14.891
",
       "      3606
",
       "      283
",
       "      2
",
       "      34
",
       "      11
",
       "      Male
",
       "      No
",
       "      Yes
",
       "      333
",
       "    
",
       "    
",
       "      1
",
       "      106.025
",
       "      6645
",
       "      483
",
       "      3
",
       "      82
",
       "      15
",
       "      Female
",
       "      Yes
",
       "      Yes
",
       "      903
",
       "    
",
       "    
",
       "      2
",
       "      104.593
",
       "      7075
",
       "      514
",
       "      4
",
       "      71
",
       "      11
",
       "      Male
",
       "      No
",
       "      No
",
       "      580
",
       "    
",
       "    
",
       "      3
",
       "      148.924
",
       "      9504
",
       "      681
",
       "      3
",
       "      36
",
       "      11
",
       "      Female
",
       "      No
",
       "      No
",
       "      964
",
       "    
",
       "    
",
       "      4
",
       "      55.882
",
       "      4897
",
       "      357
",
       "      2
",
       "      68
",
       "      16
",
       "      Male
",
       "      No
",
       "      Yes
",
       "      331
",
       "    
",
       "    
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "    
",
       "    
",
       "      395
",
       "      12.096
",
       "      4100
",
       "      307
",
       "      3
",
       "      32
",
       "      13
",
       "      Male
",
       "      No
",
       "      Yes
",
       "      560
",
       "    
",
       "    
",
       "      396
",
       "      13.364
",
       "      3838
",
       "      296
",
       "      5
",
       "      65
",
       "      17
",
       "      Male
",
       "      No
",
       "      No
",
       "      480
",
       "    
",
       "    
",
       "      397
",
       "      57.872
",
       "      4171
",
       "      321
",
       "      5
",
       "      67
",
       "      12
",
       "      Female
",
       "      No
",
       "      Yes
",
       "      138
",
       "    
",
       "    
",
       "      398
",
       "      37.728
",
       "      2525
",
       "      192
",
       "      1
",
       "      44
",
       "      13
",
       "      Male
",
       "      No
",
       "      Yes
",
       "      0
",
       "    
",
       "    
",
       "      399
",
       "      18.701
",
       "      5524
",
       "      415
",
       "      5
",
       "      64
",
       "      7
",
       "      Female
",
       "      No
",
       "      No
",
       "      966
",
       "    
",
       "  
",
       "
",
       "400 rows × 10 columns
",
       ""
      ],
      "text/plain": [
       "      Income  Limit  Rating  Cards  Age  Education  Gender Student Married  \
",
       "0     14.891   3606     283      2   34         11    Male      No     Yes   
",
       "1    106.025   6645     483      3   82         15  Female     Yes     Yes   
",
       "2    104.593   7075     514      4   71         11    Male      No      No   
",
       "3    148.924   9504     681      3   36         11  Female      No      No   
",
       "4     55.882   4897     357      2   68         16    Male      No     Yes   
",
       "..       ...    ...     ...    ...  ...        ...     ...     ...     ...   
",
       "395   12.096   4100     307      3   32         13    Male      No     Yes   
",
       "396   13.364   3838     296      5   65         17    Male      No      No   
",
       "397   57.872   4171     321      5   67         12  Female      No     Yes   
",
       "398   37.728   2525     192      1   44         13    Male      No     Yes   
",
       "399   18.701   5524     415      5   64          7  Female      No      No   
",
       "
",
       "     Balance  
",
       "0        333  
",
       "1        903  
",
       "2        580  
",
       "3        964  
",
       "4        331  
",
       "..       ...  
",
       "395      560  
",
       "396      480  
",
       "397      138  
",
       "398        0  
",
       "399      966  
",
       "
",
       "[400 rows x 10 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# read in the data into a pandas dataframe
",
    "df = pd.read_csv('CreditN400p9.csv')
",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 419
    },
    "id": "fAbLWpEI1DW8",
    "outputId": "1ef5b0ed-4ff6-4aad-b62e-80a8fa91ce08"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "
",
       "
",
       "    .dataframe tbody tr th:only-of-type {
",
       "        vertical-align: middle;
",
       "    }
",
       "
",
       "    .dataframe tbody tr th {
",
       "        vertical-align: top;
",
       "    }
",
       "
",
       "    .dataframe thead th {
",
       "        text-align: right;
",
       "    }
",
       "
",
       "
",
       "  
",
       "    
",
       "      
",
       "      Income
",
       "      Limit
",
       "      Rating
",
       "      Cards
",
       "      Age
",
       "      Education
",
       "      Gender
",
       "      Student
",
       "      Married
",
       "      Balance
",
       "    
",
       "  
",
       "  
",
       "    
",
       "      0
",
       "      14.891
",
       "      3606
",
       "      283
",
       "      2
",
       "      34
",
       "      11
",
       "      1
",
       "      0
",
       "      1
",
       "      333
",
       "    
",
       "    
",
       "      1
",
       "      106.025
",
       "      6645
",
       "      483
",
       "      3
",
       "      82
",
       "      15
",
       "      0
",
       "      1
",
       "      1
",
       "      903
",
       "    
",
       "    
",
       "      2
",
       "      104.593
",
       "      7075
",
       "      514
",
       "      4
",
       "      71
",
       "      11
",
       "      1
",
       "      0
",
       "      0
",
       "      580
",
       "    
",
       "    
",
       "      3
",
       "      148.924
",
       "      9504
",
       "      681
",
       "      3
",
       "      36
",
       "      11
",
       "      0
",
       "      0
",
       "      0
",
       "      964
",
       "    
",
       "    
",
       "      4
",
       "      55.882
",
       "      4897
",
       "      357
",
       "      2
",
       "      68
",
       "      16
",
       "      1
",
       "      0
",
       "      1
",
       "      331
",
       "    
",
       "    
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "      ...
",
       "    
",
       "    
",
       "      395
",
       "      12.096
",
       "      4100
",
       "      307
",
       "      3
",
       "      32
",
       "      13
",
       "      1
",
       "      0
",
       "      1
",
       "      560
",
       "    
",
       "    
",
       "      396
",
       "      13.364
",
       "      3838
",
       "      296
",
       "      5
",
       "      65
",
       "      17
",
       "      1
",
       "      0
",
       "      0
",
       "      480
",
       "    
",
       "    
",
       "      397
",
       "      57.872
",
       "      4171
",
       "      321
",
       "      5
",
       "      67
",
       "      12
",
       "      0
",
       "      0
",
       "      1
",
       "      138
",
       "    
",
       "    
",
       "      398
",
       "      37.728
",
       "      2525
",
       "      192
",
       "      1
",
       "      44
",
       "      13
",
       "      1
",
       "      0
",
       "      1
",
       "      0
",
       "    
",
       "    
",
       "      399
",
       "      18.701
",
       "      5524
",
       "      415
",
       "      5
",
       "      64
",
       "      7
",
       "      0
",
       "      0
",
       "      0
",
       "      966
",
       "    
",
       "  
",
       "
",
       "400 rows × 10 columns
",
       ""
      ],
      "text/plain": [
       "      Income  Limit  Rating  Cards  Age  Education  Gender  Student  Married  \
",
       "0     14.891   3606     283      2   34         11       1        0        1   
",
       "1    106.025   6645     483      3   82         15       0        1        1   
",
       "2    104.593   7075     514      4   71         11       1        0        0   
",
       "3    148.924   9504     681      3   36         11       0        0        0   
",
       "4     55.882   4897     357      2   68         16       1        0        1   
",
       "..       ...    ...     ...    ...  ...        ...     ...      ...      ...   
",
       "395   12.096   4100     307      3   32         13       1        0        1   
",
       "396   13.364   3838     296      5   65         17       1        0        0   
",
       "397   57.872   4171     321      5   67         12       0        0        1   
",
       "398   37.728   2525     192      1   44         13       1        0        1   
",
       "399   18.701   5524     415      5   64          7       0        0        0   
",
       "
",
       "     Balance  
",
       "0        333  
",
       "1        903  
",
       "2        580  
",
       "3        964  
",
       "4        331  
",
       "..       ...  
",
       "395      560  
",
       "396      480  
",
       "397      138  
",
       "398        0  
",
       "399      966  
",
       "
",
       "[400 rows x 10 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# recode the gender, student, and married preds 
",
    "df['Gender'].replace(to_replace = dict(Female = 0, Male = 1), inplace=True)
",
    "df['Student'].replace(to_replace = dict(No = 0, Yes = 1), inplace=True)
",
    "df['Married'].replace(to_replace = dict(No = 0, Yes = 1), inplace=True)
",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "id": "a6rgSVFt0-Lc",
    "outputId": "98dc7a72-11cd-4949-bbe3-073e5bac2a05"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(400, 9) (400,)
"
     ]
    }
   ],
   "source": [
    "# separate the predictors from the response
",
    "X = df.to_numpy()[:, :-1]
",
    "Y = df.to_numpy()[:, -1]
",
    "print(X.shape, Y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "id": "676w7t_e2jOr"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] Using 320 training and 80 validation samples for each CV fold.
"
     ]
    }
   ],
   "source": [
    "# instantiate the model 
",
    "elastic_net = Elastic(
",
    "    preds = X,
",
    "    responses = Y,
",
    "    lambdas = 10 ** np.arange(-2., 7.),
",
    "    alphas = [0, 0.2, 0.4, 0.6, 0.8, 1.0],
",
    "    iters = 1000,
",
    "    to_verb = True
",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "rSKcCFTQEkPz"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lambda:0.01; alpha:0; CV MSE:10234.626587559245
",
      "lambda:0.01; alpha:0.2; CV MSE:10234.211795867723
",
      "lambda:0.01; alpha:0.4; CV MSE:10233.852438234768
",
      "lambda:0.01; alpha:0.6; CV MSE:10233.436392296477
",
      "lambda:0.01; alpha:0.8; CV MSE:10233.075147217352
",
      "lambda:0.01; alpha:1.0; CV MSE:10232.692809055512
",
      "lambda:0.1; alpha:0; CV MSE:10234.620036649416
",
      "lambda:0.1; alpha:0.2; CV MSE:10230.888416232277
",
      "lambda:0.1; alpha:0.4; CV MSE:10227.51252752043
",
      "lambda:0.1; alpha:0.6; CV MSE:10224.474212258927
",
      "lambda:0.1; alpha:0.8; CV MSE:10221.717309133242
",
      "lambda:0.1; alpha:1.0; CV MSE:10219.199002422461
",
      "lambda:1.0; alpha:0; CV MSE:10234.595186864828
",
      "lambda:1.0; alpha:0.2; CV MSE:10209.723874570986
",
      "lambda:1.0; alpha:0.4; CV MSE:10200.322057813979
",
      "lambda:1.0; alpha:0.6; CV MSE:10197.62366391859
",
      "lambda:1.0; alpha:0.8; CV MSE:10198.446603910419
",
      "lambda:1.0; alpha:1.0; CV MSE:10201.418058683208
",
      "lambda:10.0; alpha:0; CV MSE:10234.556835669233
",
      "lambda:10.0; alpha:0.2; CV MSE:10233.762240772221
",
      "lambda:10.0; alpha:0.4; CV MSE:10347.53675830116
",
      "lambda:10.0; alpha:0.6; CV MSE:10513.079258829724
",
      "lambda:10.0; alpha:0.8; CV MSE:10723.196252535801
",
      "lambda:10.0; alpha:1.0; CV MSE:10972.073141404908
",
      "lambda:100.0; alpha:0; CV MSE:10234.492985310182
",
      "lambda:100.0; alpha:0.2; CV MSE:12673.274224504257
",
      "lambda:100.0; alpha:0.4; CV MSE:17043.75443270868
",
      "lambda:100.0; alpha:0.6; CV MSE:21672.861483300676
",
      "lambda:100.0; alpha:0.8; CV MSE:26104.562872343544
",
      "lambda:100.0; alpha:1.0; CV MSE:30225.406433131506
",
      "lambda:1000.0; alpha:0; CV MSE:10271.892400756773
"
     ]
    }
   ],
   "source": [
    "# fit the model
",
    "elastic_net.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "id": "AwqtUQSIOFGn",
    "outputId": "5829462d-fc37-4d4a-c8dd-1480e258fa3c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       ""
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       ""
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png":

1 CAP 5625: Programming Assignment 2 Due on Canvas by Wednesday, November 3, 2021 at 11:59pm Preliminary instructions You may consult with other students currently taking CAP 5625 at FAU on this...

Answer To: 1 CAP 5625: Programming Assignment 2 Due on Canvas by Wednesday, November 3, 2021 at 11:59pm...

Answer To This Question Is Available To Download

Related Questions & Answers

Submit New Assignment

	Income	Limit	Rating	Cards	Age	Education	Gender	Student	Married	Balance
0	14.891	3606	283	2	34	11	Male	No	Yes	333
1	106.025	6645	483	3	82	15	Female	Yes	Yes	903
2	104.593	7075	514	4	71	11	Male	No	No	580
3	148.924	9504	681	3	36	11	Female	No	No	964
4	55.882	4897	357	2	68	16	Male	No	Yes	331
...	...	...	...	...	...	...	...	...	...	...
395	12.096	4100	307	3	32	13	Male	No	Yes	560
396	13.364	3838	296	5	65	17	Male	No	No	480
397	57.872	4171	321	5	67	12	Female	No	Yes	138
398	37.728	2525	192	1	44	13	Male	No	Yes	0
399	18.701	5524	415	5	64	7	Female	No	No	966