Welcome to the second project of the Machine Learning Engineer Nanodegree! In this notebook, some template code has already been provided for you, and it will be your job to implement the additional functionality necessary to successfully complete this project. Sections that begin with 'Implementation' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. In addition, Markdown cells can be edited by typically double-clicking the cell to enter edit mode.
Answer:
This is a classification problem, because we are trying to label students with one of two labels: "needs intervention" or "doesn't need intervention".
In [ ]:
In [ ]:
In [2]:
# Import libraries
import numpy as np
import pandas as pd
from time import time
from sklearn.metrics import f1_score
# Read student data
student_data = pd.read_csv("student-data.csv")
print "Student data read successfully!"
Let's begin by investigating the dataset to determine how many students we have information on, and learn about the graduation rate among these students. In the code cell below, you will need to compute the following:
n_students.n_features.n_passed.n_failed.grad_rate, in percent (%).
In [3]:
# TODO: Calculate number of students
n_students = len(student_data)
# TODO: Calculate number of features
n_features = len(student_data.columns) - 1 # The last field is the target and is not a feature
# TODO: Calculate passing students
n_passed = len([x for x in student_data["passed"] if x == "yes"])
# TODO: Calculate failing students
n_failed = n_students - n_passed
# TODO: Calculate graduation rate
grad_rate = 100.0 * n_passed / n_students
# Print the results
print "Total number of students: {}".format(n_students)
print "Number of features: {}".format(n_features)
print "Number of students who passed: {}".format(n_passed)
print "Number of students who failed: {}".format(n_failed)
print "Graduation rate of the class: {:.2f}%".format(grad_rate)
In this section, we will prepare the data for modeling, training and testing.
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.
Run the code cell below to separate the student data into feature and target columns to see if any features are non-numeric.
In [4]:
# Extract feature columns
feature_cols = list(student_data.columns[:-1])
# Extract target column 'passed'
target_col = student_data.columns[-1]
# Show the list of columns
print "Feature columns:\n{}".format(feature_cols)
print "\nTarget column: {}".format(target_col)
# Separate the data into feature data and target data (X_all and y_all, respectively)
X_all = student_data[feature_cols]
y_all = student_data[target_col]
# Show the feature information by printing the first five rows
print "\nFeature values:"
print X_all.head()
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. Run the code cell below to perform the preprocessing routine discussed in this section.
In [5]:
def preprocess_features(X):
''' Preprocesses the student data and converts non-numeric binary variables into
binary (0/1) variables. Converts categorical variables into dummy variables. '''
# Initialize new output DataFrame
output = pd.DataFrame(index = X.index)
# Investigate each feature column for the data
for col, col_data in X.iteritems():
# If data type is non-numeric, replace all yes/no values with 1/0
if col_data.dtype == object:
col_data = col_data.replace(['yes', 'no'], [1, 0])
# If data type is categorical, convert to dummy variables
if col_data.dtype == object:
# Example: 'school' => 'school_GP' and 'school_MS'
col_data = pd.get_dummies(col_data, prefix = col)
# Collect the revised columns
output = output.join(col_data)
return output
X_all = preprocess_features(X_all)
print "Processed feature columns ({} total features):\n{}".format(len(X_all.columns), list(X_all.columns))
So far, we have converted all categorical features into numeric values. For the next step, we split the data (both features and corresponding labels) into training and test sets. In the following code cell below, you will need to implement the following:
X_all, y_all) into training and testing subsets.random_state for the function(s) you use, if provided.X_train, X_test, y_train, and y_test.
In [6]:
# TODO: Import any additional functionality you may need here
from sklearn.cross_validation import train_test_split
# TODO: Set the number of training points
num_train = 300
# Set the number of testing points
num_test = X_all.shape[0] - num_train
random_state = 0
# TODO: Shuffle and split the dataset into the number of training and testing points above
X_train, X_test, y_train, y_test = train_test_split(
X_all,
y_all,
test_size=num_test,
train_size=num_train,
random_state=random_state
)
# Show the results of the split
print "Training set has {} samples.".format(X_train.shape[0])
print "Testing set has {} samples.".format(X_test.shape[0])
In this section, you will choose 3 supervised learning models that are appropriate for this problem and available in scikit-learn. You will first discuss the reasoning behind choosing these three models by considering what you know about the data and each model's strengths and weaknesses. You will then fit the model to varying sizes of training data (100 data points, 200 data points, and 300 data points) and measure the F1 score. You will need to produce three tables (one for each model) that shows the training set size, training time, prediction time, F1 score on the training set, and F1 score on the testing set.
The following supervised learning models are currently available in scikit-learn that you may choose from:
List three supervised learning models that are appropriate for this problem. For each model chosen
Answer:
Gaussian Naive-Bayes
This method has been used in order to implement a classification algorithm in a database management system (DBMS) [1]. It was chosen because many NB operations translate well in database query languages, which enhances the scalability of the algorithm. For example, filtering and counting items is done very well by query languages, which comes in handy for calculating prior distributions of classes. Other attributes which support this solution include Gaussian NB's resilience to noise in data sets and its ability to scale well with a large amount of dimensions/features, which is likely the case in a DBMS.
The solution is demonstrated by classifying streamed JSON data from Twitter and using it to classify the emotion of the posts.
Gaussian NB is very robust when it comes to noise in the data. It also scales linearly with higher dimensionality, and can operate on discrete and continuous data. However, it does assume that features are independent, and so it does not model interrelationshiops. It also requires the data to be normally distributed.
The data we will be training on and classifying will likely have noise, and there are several dimensions, some of which are numeric and continuous, which makes NB a good candidate. However, while some of the data is normally distributed (i.e., studytime, freetime, goout), some of the data numerical data is not, and many of the features are categorical, which does not translate well to continuous distributions.
Logistic Regression
Logistic regression was used as a prediction model for customer satisfaction using factors such as on-time reliability, safety, etc. [2].
Logistic regression is ideal for splitting non-linear data. It works well when there are many examples in each category/feature, which is the case for this data.
Logistic regression would be a good candiated for this problem because the data is categorical and we are expecting a classification result. There are many examples in each category.
Ensemble Methods
Ensemble methods have been used in weather forecast systems [3]. They have led to more accurate predictions as opposed to the purely statistical post-processing systems which preceded it (and are still in use). Rather than replacing the previous models, it builds off of them, as ensemble methods are meant to do. It is also ushering in an opportunity to merge together predictions from other local forecasting systems to improve overall predictions based on a larger scope of input data.
One of the strengths of ensemble methods - specifically boosting -- is that accuracy only improves with larger training sizes, as opposed to other models which tend to start to overfit after a certain point. However, ensemble methods tend to take more computational power since they rely on multiple sub-models under the hood.
This model could make a good candidate for the student interventions system because there are many features to learn from which may or may not be independent of each other and the multiple methods of the ensemble can learn certain behaviors, much like a spam detection filter.
References
Run the code cell below to initialize three helper functions which you can use for training and testing the three supervised learning models you've chosen above. The functions are as follows:
train_classifier - takes as input a classifier and training data and fits the classifier to the data.predict_labels - takes as input a fit classifier, features, and a target labeling and makes predictions using the F1 score.train_predict - takes as input a classifier, and the training and testing data, and performs train_clasifier and predict_labels.
In [9]:
def train_classifier(clf, X_train, y_train):
''' Fits a classifier to the training data. '''
# Start the clock, train the classifier, then stop the clock
start = time()
clf.fit(X_train, y_train)
end = time()
# Print the results
print "Trained model in {:.4f} seconds".format(end - start)
def predict_labels(clf, features, target):
''' Makes predictions using a fit classifier based on F1 score. '''
# Start the clock, make predictions, then stop the clock
start = time()
y_pred = clf.predict(features)
end = time()
# Print and return results
print "Made predictions in {:.4f} seconds.".format(end - start)
return f1_score(target.values, y_pred, pos_label='yes')
def train_predict(clf, X_train, y_train, X_test, y_test):
''' Train and predict using a classifer based on F1 score. '''
# Indicate the classifier and the training set size
print "Training a {} using a training set size of {}. . .".format(clf.__class__.__name__, len(X_train))
# Train the classifier
train_classifier(clf, X_train, y_train)
# Print the results of prediction for both training and testing
print "F1 score for training set: {:.4f}.".format(predict_labels(clf, X_train, y_train))
print "F1 score for test set: {:.4f}.".format(predict_labels(clf, X_test, y_test))
With the predefined functions above, you will now import the three supervised learning models of your choice and run the train_predict function for each one. Remember that you will need to train and predict on each classifier for three different training set sizes: 100, 200, and 300. Hence, you should expect to have 9 different outputs below — 3 for each model using the varying training set sizes. In the following code cell, you will need to implement the following:
clf_A, clf_B, and clf_C.random_state for each model you use, if provided.X_train and y_train.
In [8]:
# TODO: Import the three supervised learning models from sklearn
from sklearn.naive_bayes import GaussianNB
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import AdaBoostClassifier
# TODO: Initialize the three models
clf_A = GaussianNB()
clf_B = LogisticRegression(random_state=14)
clf_C = AdaBoostClassifier(random_state=14)
# TODO: Set up the training set sizes
X_train_100 = X_train[:100]
y_train_100 = y_train[:100]
X_train_200 = X_train[:200]
y_train_200 = y_train[:200]
X_train_300 = X_train[:300]
y_train_300 = y_train[:300]
# TODO: Execute the 'train_predict' function for each classifier and each training set size
# train_predict(clf, X_train, y_train, X_test, y_test)
for clf in [clf_A, clf_B, clf_C]:
for X_train_N, y_train_N in [(X_train_100, y_train_100), (X_train_200, y_train_200), (X_train_300, y_train_300)]:
train_predict(clf, X_train_N, y_train_N, X_test, y_test)
print("")
Edit the cell below to see how a table can be designed in Markdown. You can record your results from above in the tables provided.
Classifer 1 - Gaussian Naive Bayes
| Training Set Size | Training Time | Prediction Time (test) | F1 Score (train) | F1 Score (test) |
|---|---|---|---|---|
| 100 | 0.0020 | 0.0000 | 0.8550 | 0.7481 |
| 200 | 0.0010 | 0.0000 | 0.8321 | 0.7132 |
| 300 | 0.0020 | 0.0000 | 0.8088 | 0.7500 |
Classifer 2 - Logistic Regression
| Training Set Size | Training Time | Prediction Time (test) | F1 Score (train) | F1 Score (test) |
|---|---|---|---|---|
| 100 | 0.0020 | 0.0000 | 0.8571 | 0.7612 |
| 200 | 0.0020 | 0.0000 | 0.8380 | 0.7794 |
| 300 | 0.0040 | 0.0000 | 0.8381 | 0.7910 |
Classifer 3 - AdaBoost
| Training Set Size | Training Time | Prediction Time (test) | F1 Score (train) | F1 Score (test) |
|---|---|---|---|---|
| 100 | 0.0920 | 0.0160 | 0.9538 | 0.7200 |
| 200 | 0.1220 | 0.0060 | 0.8826 | 0.8058 |
| 300 | 0.1120 | 0.0160 | 0.8688 | 0.7794 |
In this final section, you will choose from the three supervised learning models the best model to use on the student data. You will then perform a grid search optimization for the model over the entire training set (X_train and y_train) by tuning at least one parameter to improve upon the untuned model's F1 score.
Based on the experiments you performed earlier, in one to two 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?
Answer:
Based on my experiments, I chose to use logistic regression as the model for the student intervention system. Of the three models chosen, this model fits both computational and accuracy requirements the best. Logisitic regression performed consistently well with the given training/test data set, but was noticeably quicker for both the training and prediction time cmopared to the similarly accuracte Adaboost classifier. The Gaussian Naive Bayes classifier was similar in speed to the logistic regression classifier, but was not as accurate. Logsitic regression is the best model given the available data and requirements.
In one to two paragraphs, explain to the board of directors in layman's terms how the final model chosen is supposed to work. Be sure that you are describing the major qualities of the model, such as how the model is trained and how the model makes a prediction. Avoid using advanced mathematical or technical jargon, such as describing equations or discussing the algorithm implementation.
Answer:
Logistic regression uses a function to map input features (e.g. absences, free time, family size, etc.) to one of a set of labels (e.g. "requires intervention" or "doesn't require intervention"). The function is based off of the probability of that label occurring given that the feature occurred. The particular function used is bound by 0 and 1, where values less than 0.5 are associated with one label, and values above 0.5 are associated with the other label.
During the training process, the probabilities for the function are estimated based on the occurrence rate in the training data. In order to account for unequal contributions of the features, certain weights are assigned to each feature. The weights are found using an algorithm that maximizes weights for examples which "help" find positive results (i.e. the prediction matches the outcome), and minimize the weights of those that do not.
Once all of the weights have been calculated, new sets of features may be given to the resulting function to output whether or not the student requires intervention. Because the mechanism for classifying the student is based on a simple mathematical function, as opposed to other models that use neighboring points or node traversal, runtime is very computationally and memory efficient.
Fine tune the chosen model. Use grid search (GridSearchCV) with at least one important parameter tuned with at least 3 different values. You will need to use the entire training set for this. In the code cell below, you will need to implement the following:
sklearn.grid_search.GridSearchCV and sklearn.metrics.make_scorer.parameters = {'parameter' : [list of values]}.clf.make_scorer and store it in f1_scorer.pos_label parameter to the correct value!clf using f1_scorer as the scoring method, and store it in grid_obj.X_train, y_train), and store it in grid_obj.
In [10]:
# TODO: Import 'GridSearchCV' and 'make_scorer'
from sklearn.grid_search import GridSearchCV
from sklearn.metrics import make_scorer
# TODO: Create the parameters list you wish to tune
num_features = len(feature_cols)
parameters = {
"C": [0.5, 1.0, 1.5, 2.0]
}
# TODO: Initialize the classifier
clf = LogisticRegression(random_state=14)
# TODO: Make an f1 scoring function using 'make_scorer'
f1_scorer = make_scorer(f1_score, pos_label="yes")
# TODO: Perform grid search on the classifier using the f1_scorer as the scoring method
grid_obj = GridSearchCV(clf, param_grid=parameters, scoring = f1_scorer)
# TODO: Fit the grid search object to the training data and find the optimal parameters
grid_obj = grid_obj.fit(X_train, y_train)
# Get the estimator
clf = grid_obj.best_estimator_
# Report the final F1 score for training and testing after parameter tuning
print "Tuned model has a training F1 score of {:.4f}.".format(predict_labels(clf, X_train, y_train))
print "Tuned model has a testing F1 score of {:.4f}.".format(predict_labels(clf, X_test, y_test))
Answer:
Training F1 score: 0.8363
Test F1 score: 0.8000
These are similar to the untuned model, but the test score is a little better, which is good because it means that the model improved accuracy without losing its ability to generalize.
Note: Once you have completed all of the code implementations and successfully answered each question above, you may finalize your work by exporting the iPython Notebook as an HTML document. You can do this by using the menu above and navigating to
File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.