Project 2: Supervised Learning

Building a Student Intervention System

1. Classification vs Regression

Your goal is to identify students who might need early intervention - which type of supervised machine learning problem is this, classification or regression? Why?

2. Exploring the Data

Let's go ahead and read in the student dataset first.

To execute a code cell, click inside it and press Shift+Enter.


In [1]:
# Import libraries
import numpy as np
import pandas as pd

In [2]:
# Read student data
student_data = pd.read_csv("student-data.csv")
print "Student data read successfully!"
# Note: The last column 'passed' is the target/label, all other are feature columns
#student_data.head()


Student data read successfully!

Now, can you find out the following facts about the dataset?

  • Total number of students
  • Number of students who passed
  • Number of students who failed
  • Graduation rate of the class (%)
  • Number of features

Use the code block below to compute these values. Instructions/steps are marked using TODOs.


In [3]:
# TODO: Compute desired values - replace each '?' with an appropriate expression/function call
n_students = student_data.shape[0]
n_features = student_data.shape[1]-1
n_passed   = student_data.loc[student_data['passed']=='yes','passed'].count()
n_failed   = student_data.loc[student_data['passed']=='no','passed'].count()
grad_rate  = (n_passed*100.0)/(n_passed+n_failed)
print "Total number of students: {}".format(n_students)
print "Number of students who passed: {}".format(n_passed)
print "Number of students who failed: {}".format(n_failed)
print "Number of features: {}".format(n_features)
print "Graduation rate of the class: {:.2f}%".format(grad_rate)


Total number of students: 395
Number of students who passed: 265
Number of students who failed: 130
Number of features: 30
Graduation rate of the class: 67.09%

3. Preparing the Data

In this section, we will prepare the data for modeling, training and testing.

Identify feature and target columns

It is often the case that the data you obtain contains non-numeric features. This can be a problem, as most machine learning algorithms expect numeric data to perform computations with.

Let's first separate our data into feature and target columns, and see if any features are non-numeric.
Note: For this dataset, the last column ('passed') is the target or label we are trying to predict.


In [4]:
# Extract feature (X) and target (y) columns
feature_cols = list(student_data.columns[:-1])  # all columns but last are features
target_col = student_data.columns[-1]  # last column is the target/label
print "Feature column(s):-\n{}".format(feature_cols)
print "Target column: {}".format(target_col)

X_all = student_data[feature_cols]  # feature values for all students
y_all = student_data[target_col]  # corresponding targets/labels
print "\nFeature values:-"
print X_all.head()  # print the first 5 rows


Feature column(s):-
['school', 'sex', 'age', 'address', 'famsize', 'Pstatus', 'Medu', 'Fedu', 'Mjob', 'Fjob', 'reason', 'guardian', 'traveltime', 'studytime', 'failures', 'schoolsup', 'famsup', 'paid', 'activities', 'nursery', 'higher', 'internet', 'romantic', 'famrel', 'freetime', 'goout', 'Dalc', 'Walc', 'health', 'absences']
Target column: passed

Feature values:-
  school sex  age address famsize Pstatus  Medu  Fedu     Mjob      Fjob  \
0     GP   F   18       U     GT3       A     4     4  at_home   teacher   
1     GP   F   17       U     GT3       T     1     1  at_home     other   
2     GP   F   15       U     LE3       T     1     1  at_home     other   
3     GP   F   15       U     GT3       T     4     2   health  services   
4     GP   F   16       U     GT3       T     3     3    other     other   

    ...    higher internet  romantic  famrel  freetime goout Dalc Walc health  \
0   ...       yes       no        no       4         3     4    1    1      3   
1   ...       yes      yes        no       5         3     3    1    1      3   
2   ...       yes      yes        no       4         3     2    2    3      3   
3   ...       yes      yes       yes       3         2     2    1    1      5   
4   ...       yes       no        no       4         3     2    1    2      5   

  absences  
0        6  
1        4  
2       10  
3        2  
4        4  

[5 rows x 30 columns]

Preprocess feature columns

As you can see, there are several non-numeric columns that need to be converted! Many of them are simply yes/no, e.g. internet. These can be reasonably converted into 1/0 (binary) values.

Other columns, like Mjob and Fjob, have more than two values, and are known as categorical variables. The recommended way to handle such a column is to create as many columns as possible values (e.g. Fjob_teacher, Fjob_other, Fjob_services, etc.), and assign a 1 to one of them and 0 to all others.

These generated columns are sometimes called dummy variables, and we will use the pandas.get_dummies() function to perform this transformation.


In [5]:
# Preprocess feature columns
def preprocess_features(X):
    outX = pd.DataFrame(index=X.index)  # output dataframe, initially empty

    # Check each column
    for col, col_data in X.iteritems():
        # If data type is non-numeric, try to replace all yes/no values with 1/0
        if col_data.dtype == object:
            col_data = col_data.replace(['yes', 'no'], [1, 0])
        # Note: This should change the data type for yes/no columns to int

        # If still non-numeric, convert to one or more dummy variables
        if col_data.dtype == object:
            col_data = pd.get_dummies(col_data, prefix=col)  # e.g. 'school' => 'school_GP', 'school_MS'

        outX = outX.join(col_data)  # collect column(s) in output dataframe

    return outX

X_all = preprocess_features(X_all)
print "Processed feature columns ({}):-\n{}".format(len(X_all.columns), list(X_all.columns))


Processed feature columns (48):-
['school_GP', 'school_MS', 'sex_F', 'sex_M', 'age', 'address_R', 'address_U', 'famsize_GT3', 'famsize_LE3', 'Pstatus_A', 'Pstatus_T', 'Medu', 'Fedu', 'Mjob_at_home', 'Mjob_health', 'Mjob_other', 'Mjob_services', 'Mjob_teacher', 'Fjob_at_home', 'Fjob_health', 'Fjob_other', 'Fjob_services', 'Fjob_teacher', 'reason_course', 'reason_home', 'reason_other', 'reason_reputation', 'guardian_father', 'guardian_mother', 'guardian_other', 'traveltime', 'studytime', 'failures', 'schoolsup', 'famsup', 'paid', 'activities', 'nursery', 'higher', 'internet', 'romantic', 'famrel', 'freetime', 'goout', 'Dalc', 'Walc', 'health', 'absences']

Split data into training and test sets

So far, we have converted all categorical features into numeric values. In this next step, we split the data (both features and corresponding labels) into training and test sets.


In [6]:
# First, decide how many training vs test samples you want
num_all = student_data.shape[0]  # same as len(student_data)
num_train = 300  # about 75% of the data
num_test = num_all - num_train

# TODO: Then, select features (X) and corresponding labels (y) for the training and test sets
# Note: Shuffle the data or randomly select samples to avoid any bias due to ordering in the dataset
from sklearn import cross_validation

X_train,X_test,y_train,y_test=cross_validation.train_test_split(X_all,y_all,train_size=num_train,random_state=100)
print "Training set: {} samples".format(X_train.shape[0])
print "Test set: {} samples".format(X_test.shape[0])
# Note: If you need a validation set, extract it from within training data


Training set: 300 samples
Test set: 95 samples

4. Training and Evaluating Models

Choose 3 supervised learning models that are available in scikit-learn, and appropriate for this problem. For each model:

  • What are the general applications of this model? What are its strengths and weaknesses?
  • Given what you know about the data so far, why did you choose this model to apply?
  • Fit this model to the training data, try to predict labels (for both training and test sets), and measure the F1 score. Repeat this process with different training set sizes (100, 200, 300), keeping test set constant.

Produce a table showing training time, prediction time, F1 score on training set and F1 score on test set, for each training set size.

Note: You need to produce 3 such tables - one for each model.


In [7]:
# Train a model
import time

def train_classifier(clf, X_train, y_train):
    print "Training {}...".format(clf.__class__.__name__)
    start = time.time()
    clf.fit(X_train, y_train)
    end = time.time()
    print "Done!\nTraining time (secs): {:.3f}".format(end - start)

In [8]:
# Predict on training set and compute F1 score
from sklearn.metrics import f1_score

def predict_labels(clf, features, target):
    print "Predicting labels using {}...".format(clf.__class__.__name__)
    start = time.time()
    y_pred = clf.predict(features)
    end = time.time()
    print "Done!\nPrediction time (secs): {:.3f}".format(end - start)
    return f1_score(target.values, y_pred, pos_label='yes')

In [9]:
# Train and predict using different training set sizes
def train_predict(clf, X_train, y_train, X_test, y_test):
    print "------------------------------------------"
    print "Training set size: {}".format(len(X_train))
    train_classifier(clf, X_train, y_train)
    print "F1 score for training set: {}".format(predict_labels(clf, X_train, y_train))
    print "F1 score for test set: {}".format(predict_labels(clf, X_test, y_test))

In [10]:
# TODO: Choose a model, import it and instantiate an object
from sklearn.naive_bayes import GaussianNB
from sklearn.tree import DecisionTreeClassifier
from sklearn import svm

In [11]:
clf_NB = GaussianNB()
clf_DT = DecisionTreeClassifier(min_samples_split=5)
clf_SV = svm.SVC()

In [12]:
for clf in [clf_NB,clf_DT,clf_SV]:
    for size in [100,200,300]:
        train_predict(clf,X_train[:size],y_train[:size],X_test,y_test)


------------------------------------------
Training set size: 100
Training GaussianNB...
Done!
Training time (secs): 0.003
Predicting labels using GaussianNB...
Done!
Prediction time (secs): 0.001
F1 score for training set: 0.494117647059
Predicting labels using GaussianNB...
Done!
Prediction time (secs): 0.001
F1 score for test set: 0.408602150538
------------------------------------------
Training set size: 200
Training GaussianNB...
Done!
Training time (secs): 0.002
Predicting labels using GaussianNB...
Done!
Prediction time (secs): 0.001
F1 score for training set: 0.810606060606
Predicting labels using GaussianNB...
Done!
Prediction time (secs): 0.001
F1 score for test set: 0.75
------------------------------------------
Training set size: 300
Training GaussianNB...
Done!
Training time (secs): 0.002
Predicting labels using GaussianNB...
Done!
Prediction time (secs): 0.001
F1 score for training set: 0.785542168675
Predicting labels using GaussianNB...
Done!
Prediction time (secs): 0.001
F1 score for test set: 0.791044776119
------------------------------------------
Training set size: 100
Training DecisionTreeClassifier...
Done!
Training time (secs): 0.047
Predicting labels using DecisionTreeClassifier...
Done!
Prediction time (secs): 0.000
F1 score for training set: 0.976
Predicting labels using DecisionTreeClassifier...
Done!
Prediction time (secs): 0.000
F1 score for test set: 0.633333333333
------------------------------------------
Training set size: 200
Training DecisionTreeClassifier...
Done!
Training time (secs): 0.002
Predicting labels using DecisionTreeClassifier...
Done!
Prediction time (secs): 0.000
F1 score for training set: 0.961832061069
Predicting labels using DecisionTreeClassifier...
Done!
Prediction time (secs): 0.000
F1 score for test set: 0.661290322581
------------------------------------------
Training set size: 300
Training DecisionTreeClassifier...
Done!
Training time (secs): 0.002
Predicting labels using DecisionTreeClassifier...
Done!
Prediction time (secs): 0.000
F1 score for training set: 0.961038961039
Predicting labels using DecisionTreeClassifier...
Done!
Prediction time (secs): 0.000
F1 score for test set: 0.763358778626
------------------------------------------
Training set size: 100
Training SVC...
Done!
Training time (secs): 0.002
Predicting labels using SVC...
Done!
Prediction time (secs): 0.001
F1 score for training set: 0.882352941176
Predicting labels using SVC...
Done!
Prediction time (secs): 0.001
F1 score for test set: 0.774647887324
------------------------------------------
Training set size: 200
Training SVC...
Done!
Training time (secs): 0.004
Predicting labels using SVC...
Done!
Prediction time (secs): 0.003
F1 score for training set: 0.881355932203
Predicting labels using SVC...
Done!
Prediction time (secs): 0.002
F1 score for test set: 0.783783783784
------------------------------------------
Training set size: 300
Training SVC...
Done!
Training time (secs): 0.009
Predicting labels using SVC...
Done!
Prediction time (secs): 0.007
F1 score for training set: 0.845986984816
Predicting labels using SVC...
Done!
Prediction time (secs): 0.003
F1 score for test set: 0.833333333333

5. Choosing the Best Model

  • Based on the experiments you performed earlier, in 1-2 paragraphs explain to the board of supervisors what single model you chose as the best model. Which model is generally the most appropriate based on the available data, limited resources, cost, and performance?
  • In 1-2 paragraphs explain to the board of supervisors in layman's terms how the final model chosen is supposed to work (for example if you chose a Decision Tree or Support Vector Machine, how does it make a prediction).
  • Fine-tune the model. Use Gridsearch with at least one important parameter tuned and with at least 3 settings. Use the entire training set for this.
  • What is the model's final F1 score?

In [13]:
from sklearn.grid_search import GridSearchCV 
from sklearn.cross_validation import StratifiedShuffleSplit
from sklearn.metrics import f1_score
from sklearn.metrics import make_scorer

f1_scorer = make_scorer(f1_score, pos_label="yes")

parameters=  {'gamma': [1e-2, 1e-4, 1e-6], 
              'C': [10, 100, 200, 300, 500, 700],
              'kernel':['rbf','sigmoid','poly']
             }  

ssscv = StratifiedShuffleSplit( y_train, n_iter=10, test_size=0.1) 

grid = GridSearchCV( svm.SVC(), parameters, cv = ssscv , scoring=f1_scorer) 

grid.fit( X_train, y_train ) 
best = grid.best_estimator_ 
y_pred = best.predict( X_test ) 

print "F1 score: {}".format( f1_score( y_test, y_pred, pos_label = 'yes' ))
print "Best params: {}".format( grid.best_params_ )


F1 score: 0.859060402685
Best params: {'kernel': 'rbf', 'C': 200, 'gamma': 0.0001}