Machine Learning Engineer Nanodegree¶
Supervised Learning¶
Project: Finding Donors for CharityML¶
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.
Getting Started¶
In this project, you will employ several supervised algorithms of your choice to accurately model individuals' income using data collected from the 1994 U.S. Census. You will then choose the best candidate algorithm from preliminary results and further optimize this algorithm to best model the data. Your goal with this implementation is to construct a model that accurately predicts whether an individual makes more than $50,000. This sort of task can arise in a non-profit setting, where organizations survive on donations. Understanding an individual's income can help a non-profit better understand how large of a donation to request, or whether or not they should reach out to begin with. While it can be difficult to determine an individual's general income bracket directly from public sources, we can (as we will see) infer this value from other publically available features.
The dataset for this project originates from the UCI Machine Learning Repository. The datset was donated by Ron Kohavi and Barry Becker, after being published in the article "Scaling Up the Accuracy of Naive-Bayes Classifiers: A Decision-Tree Hybrid". You can find the article by Ron Kohavi online. The data we investigate here consists of small changes to the original dataset, such as removing the 'fnlwgt' feature and records with missing or ill-formatted entries.
Exploring the Data¶
Run the code cell below to load necessary Python libraries and load the census data. Note that the last column from this dataset, 'income', will be our target label (whether an individual makes more than, or at most, $50,000 annually). All other columns are features about each individual in the census database.
# Import libraries necessary for this project
import numpy as np
import pandas as pd
from time import time
from IPython.display import display # Allows the use of display() for DataFrames
# Import supplementary visualization code visuals.py
import visuals as vs
# Pretty display for notebooks
%matplotlib inline
# Load the Census dataset
data = pd.read_csv("census.csv")
# Success - Display the first record
display(data.head(n=30))
Implementation: Data Exploration¶
A cursory investigation of the dataset will determine how many individuals fit into either group, and will tell us about the percentage of these individuals making more than \$50,000. In the code cell below, you will need to compute the following:
- The total number of records,
'n_records' - The number of individuals making more than \$50,000 annually,
'n_greater_50k'. - The number of individuals making at most \$50,000 annually,
'n_at_most_50k'. - The percentage of individuals making more than \$50,000 annually,
'greater_percent'.
Hint: You may need to look at the table above to understand how the 'income' entries are formatted.
df = pd.DataFrame(data)
# TODO: Total number of records
n_records = len(df.index)
# TODO: Number of records where individual's income is more than $50,000
n_greater_50k = len(df[df.income == ">50K"])
# TODO: Number of records where individual's income is at most $50,000
n_at_most_50k = len(df[df.income == "<=50K"])
# TODO: Percentage of individuals whose income is more than $50,000
greater_percent = float(n_greater_50k) / float(n_records) * 100
# Print the results
print 'Total number of records: {}'.format(n_records)
print 'Individuals making more than $50,000: {}'.format(n_greater_50k)
print 'Individuals making at most $50,000: {}'.format(n_at_most_50k)
print 'Percentage of individuals making more than $50,000: {:.2f}%'.format(greater_percent)
Preparing the Data¶
Before data can be used as input for machine learning algorithms, it often must be cleaned, formatted, and restructured — this is typically known as preprocessing. Fortunately, for this dataset, there are no invalid or missing entries we must deal with, however, there are some qualities about certain features that must be adjusted. This preprocessing can help tremendously with the outcome and predictive power of nearly all learning algorithms.
Transforming Skewed Continuous Features¶
A dataset may sometimes contain at least one feature whose values tend to lie near a single number, but will also have a non-trivial number of vastly larger or smaller values than that single number. Algorithms can be sensitive to such distributions of values and can underperform if the range is not properly normalized. With the census dataset two features fit this description: 'capital-gain' and 'capital-loss'.
Run the code cell below to plot a histogram of these two features. Note the range of the values present and how they are distributed.
# Split the data into features and target label
income_raw = data['income']
features_raw = data.drop('income', axis = 1)
# Visualize skewed continuous features of original data
vs.distribution(data)
For highly-skewed feature distributions such as 'capital-gain' and 'capital-loss', it is common practice to apply a logarithmic transformation on the data so that the very large and very small values do not negatively affect the performance of a learning algorithm. Using a logarithmic transformation significantly reduces the range of values caused by outliers. Care must be taken when applying this transformation however: The logarithm of 0 is undefined, so we must translate the values by a small amount above 0 to apply the the logarithm successfully.
Run the code cell below to perform a transformation on the data and visualize the results. Again, note the range of values and how they are distributed.
# Log-transform the skewed features
skewed = ['capital-gain', 'capital-loss']
features_raw[skewed] = data[skewed].apply(lambda x: np.log(x + 1))
# Visualize the new log distributions
vs.distribution(features_raw, transformed = True)
Normalizing Numerical Features¶
In addition to performing transformations on features that are highly skewed, it is often good practice to perform some type of scaling on numerical features. Applying a scaling to the data does not change the shape of each feature's distribution (such as 'capital-gain' or 'capital-loss' above); however, normalization ensures that each feature is treated equally when applying supervised learners. Note that once scaling is applied, observing the data in its raw form will no longer have the same original meaning, as exampled below.
Run the code cell below to normalize each numerical feature. We will use sklearn.preprocessing.MinMaxScaler for this.
# Import sklearn.preprocessing.StandardScaler
from sklearn.preprocessing import MinMaxScaler
# Initialize a scaler, then apply it to the features
scaler = MinMaxScaler()
numerical = ['age', 'education-num', 'capital-gain', 'capital-loss', 'hours-per-week']
features_raw[numerical] = scaler.fit_transform(data[numerical])
# Show an example of a record with scaling applied
display(features_raw.head(n = 8))
Implementation: Data Preprocessing¶
From the table in Exploring the Data above, we can see there are several features for each record that are non-numeric. Typically, learning algorithms expect input to be numeric, which requires that non-numeric features (called categorical variables) be converted. One popular way to convert categorical variables is by using the one-hot encoding scheme. One-hot encoding creates a "dummy" variable for each possible category of each non-numeric feature. For example, assume someFeature has three possible entries: A, B, or C. We then encode this feature into someFeature_A, someFeature_B and someFeature_C.
| someFeature | someFeature_A | someFeature_B | someFeature_C | ||
|---|---|---|---|---|---|
| 0 | B | 0 | 1 | 0 | |
| 1 | C | ----> one-hot encode ----> | 0 | 0 | 1 |
| 2 | A | 1 | 0 | 0 |
Additionally, as with the non-numeric features, we need to convert the non-numeric target label, 'income' to numerical values for the learning algorithm to work. Since there are only two possible categories for this label ("<=50K" and ">50K"), we can avoid using one-hot encoding and simply encode these two categories as 0 and 1, respectively. In code cell below, you will need to implement the following:
- Use
pandas.get_dummies()to perform one-hot encoding on the'features_raw'data. - Convert the target label
'income_raw'to numerical entries.- Set records with "<=50K" to
0and records with ">50K" to1.
- Set records with "<=50K" to
# TODO: One-hot encode the 'features_raw' data using pandas.get_dummies()
features = pd.get_dummies(features_raw)
# TODO: Encode the 'income_raw' data to numerical values
income = income_raw.apply(lambda x: 1 if x == '>50K' else 0)
# Print the number of features after one-hot encoding
encoded = list(features.columns)
print "{} total features after one-hot encoding.".format(len(encoded))
# Uncomment the following line to see the encoded feature names
print encoded
import seaborn as sns
sns.set(style="whitegrid", color_codes=True)
sns.factorplot("sex", col="education_level", data=data,
hue='income', kind="count", col_wrap=4);
Shuffle and Split Data¶
Now all categorical variables have been converted into numerical features, and all numerical features have been normalized. As always, we will now split the data (both features and their labels) into training and test sets. 80% of the data will be used for training and 20% for testing.
Run the code cell below to perform this split.
# Import train_test_split
from sklearn.cross_validation import train_test_split
# Split the 'features' and 'income' data into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(features, income, test_size=0.2, random_state = 0)
# 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])
Evaluating Model Performance¶
In this section, we will investigate four different algorithms, and determine which is best at modeling the data. Three of these algorithms will be supervised learners of your choice, and the fourth algorithm is known as a naive predictor.
Metrics and the Naive Predictor¶
CharityML, equipped with their research, knows individuals that make more than \$50,000 are most likely to donate to their charity. Because of this, *CharityML* is particularly interested in predicting who makes more than \$50,000 accurately. It would seem that using accuracy as a metric for evaluating a particular model's performace would be appropriate. Additionally, identifying someone that does not make more than \$50,000 as someone who does would be detrimental to *CharityML*, since they are looking to find individuals willing to donate. Therefore, a model's ability to precisely predict those that make more than \$50,000 is more important than the model's ability to recall those individuals. We can use F-beta score as a metric that considers both precision and recall:
$$ F_{\beta} = (1 + \beta^2) \cdot \frac{precision \cdot recall}{\left( \beta^2 \cdot precision \right) + recall} $$In particular, when $\beta = 0.5$, more emphasis is placed on precision. This is called the F$_{0.5}$ score (or F-score for simplicity).
Looking at the distribution of classes (those who make at most \$50,000, and those who make more), it's clear most individuals do not make more than \$50,000. This can greatly affect accuracy, since we could simply say "this person does not make more than \$50,000" and generally be right, without ever looking at the data! Making such a statement would be called naive, since we have not considered any information to substantiate the claim. It is always important to consider the naive prediction for your data, to help establish a benchmark for whether a model is performing well. That been said, using that prediction would be pointless: If we predicted all people made less than \$50,000, CharityML would identify no one as donors.
Question 1 - Naive Predictor Performace¶
If we chose a model that always predicted an individual made more than \$50,000, what would that model's accuracy and F-score be on this dataset?
Note: You must use the code cell below and assign your results to 'accuracy' and 'fscore' to be used later.
# TODO: Calculate accuracy
accuracy = float(11208) / float(45222)
# TODO: Calculate F-score using the formula above for beta = 0.5
fscore = float((1+.25)*(float(11208)/float(45222*1)))/float(((.25 * float(11208)/float(45222)) + 1))
# Print the results
print "Naive Predictor: [Accuracy score: {:.4f}, F-score: {:.4f}]".format(accuracy, fscore)
Supervised Learning Models¶
The following supervised learning models are currently available in scikit-learn that you may choose from:
- Gaussian Naive Bayes (GaussianNB)
- Decision Trees
- Ensemble Methods (Bagging, AdaBoost, Random Forest, Gradient Boosting)
- K-Nearest Neighbors (KNeighbors)
- Stochastic Gradient Descent Classifier (SGDC)
- Support Vector Machines (SVM)
- Logistic Regression
Question 2 - Model Application¶
List three of the supervised learning models above that are appropriate for this problem that you will test on the census data. For each model chosen
- Describe one real-world application in industry where the model can be applied. (You may need to do research for this — give references!)
- What are the strengths of the model; when does it perform well?
- What are the weaknesses of the model; when does it perform poorly?
- What makes this model a good candidate for the problem, given what you know about the data?
Answer:
Ensemble Methods (Bagging, AdaBoost, Random Forest, Gradient Boosting)
• Describe one real-world application in industry where the model can be applied.
Ensemble Methods are particularly good at predictive tasks, and due to their general purpose nature can be applied to many different situations. One real-world example of their application is making product recommendations to users, a specific example being the application of a gradient boosted decision tree model in the Netflix Prize competition on Kaggle (http://www.netflixprize.com/assets/GrandPrize2009_BPC_BellKor.pdf).
• What are the strengths of the model; when does it perform well?
The major strength of Ensemble Methods is that it is very difficult for the model to overfit, because the model only improves as you increase the complexity of the model (e.g. Bagging; by building more and more independent estimators, the resulting average of the estimators will only improve).
• What are the weaknesses of the model; when does it perform poorly?
The major weaknesses to do with Ensemble Methods are related to increased storage and computation. Since Ensemble Methods require multiple classifiers, there is a need for more storage for the classifiers which can be costly. In a simliar vein, all of the classifiers need to be processed, rather than just one, which means that the run time of the algorithm is higher.
• What makes this model a good candidate for the problem, given what you know about the data?
Since the task is to construct a model to perform binary classification of whether an individual makes over $50k or not, Ensemble Methods should generate valuable results.
Decision Trees
• Describe one real-world application in industry where the model can be applied.
One real-world application of Decision Trees is in the 'Automated Identification of Cosmic-Ray Hits in Hubble Space Telescope Images' (http://adsabs.harvard.edu/full/1995PASP..107..279S). The goal of the model is to predict cosmic-ray hits, and has successfully done so with 95% accuracy. Decision Trees were well suited to the problem because of the need for fast classification time.
• What are the strengths of the model; when does it perform well?
The major strengths of Decision Tree models are that they are easy to use, run quickly, able to handle both categorical & numerical data, and graphically allow you to interpret the data.
• What are the weaknesses of the model; when does it perform poorly?
The major weakness of Decision Trees is that it is highly prone to overfitting. One solution to this is to track its performance and prune the branches so that not too many features are included.
• What makes this model a good candidate for the problem, given what you know about the data?
A Decision Tree model is a good candidate for this problem as Decision Tree models are particularly adept at binary classification, however it may run into problems due to the number of features (especially all the iterations of native-country) so care will have to be taken with regards to feature selection.
Support Vector Machines (SVM)
• Describe one real-world application in industry where the model can be applied.
One real-world application of Support Vector Machines is in modelling the prediction for common diseases, in this particular case diabetes (https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/1472-6947-10-16). The SVM was used to select the best set of variables to classify individuals into categories relating to whether they had different stages of diabetes (Classifcation Scheme 1 being diagnosed or undiagnosed diabetes vs. pre-diabetes or no diabetes and Classification Scheme II being undiagnosed diabetes or pre-diabetes vs. no diabetes).
• What are the strengths of the model; when does it perform well?
SVMs work particularly well in complicated domains where there's a clear margin of separation.
• What are the weaknesses of the model; when does it perform poorly?
SVMs don't work well in large datasets because of the growing amount of time it takes to classify the data. They also don't work well when there's lots of noise, as SVMs can overfit to noise in the data.
• What makes this model a good candidate for the problem, given what you know about the data?
SVMs are strongly capable of classifying non-linear data which may prove useful in this dataset, but the main reason I chose the model was because I am curious to find out if this dataset is too large and has too many features for an SVM to perform successfully. In as valuable as it is to find the right models, I think it is equally valuable to prove to yourself that certain models may not be the best choice.
Implementation - Creating a Training and Predicting Pipeline¶
To properly evaluate the performance of each model you've chosen, it's important that you create a training and predicting pipeline that allows you to quickly and effectively train models using various sizes of training data and perform predictions on the testing data. Your implementation here will be used in the following section. In the code block below, you will need to implement the following:
- Import
fbeta_scoreandaccuracy_scorefromsklearn.metrics. - Fit the learner to the sampled training data and record the training time.
- Perform predictions on the test data
X_test, and also on the first 300 training pointsX_train[:300].- Record the total prediction time.
- Calculate the accuracy score for both the training subset and testing set.
- Calculate the F-score for both the training subset and testing set.
- Make sure that you set the
betaparameter!
- Make sure that you set the
# TODO: Import two metrics from sklearn - fbeta_score and accuracy_score
from sklearn.metrics import fbeta_score
from sklearn.metrics import accuracy_score
def train_predict(learner, sample_size, X_train, y_train, X_test, y_test):
'''
inputs:
- learner: the learning algorithm to be trained and predicted on
- sample_size: the size of samples (number) to be drawn from training set
- X_train: features training set
- y_train: income training set
- X_test: features testing set
- y_test: income testing set
'''
results = {}
# TODO: Fit the learner to the training data using slicing with 'sample_size'
start = time() # Get start time
learner = learner.fit(X_train[:sample_size], y_train[:sample_size])
end = time() # Get end time
# TODO: Calculate the training time
results['train_time'] = end - start
# TODO: Get the predictions on the test set,
# then get predictions on the first 300 training samples
start = time() # Get start time
predictions_test = learner.predict(X_test)
predictions_train = learner.predict(X_train[:300])
end = time() # Get end time
# TODO: Calculate the total prediction time
results['pred_time'] = end - start
# TODO: Compute accuracy on the first 300 training samples
results['acc_train'] = accuracy_score(y_train[:300], predictions_train)
# TODO: Compute accuracy on test set
results['acc_test'] = accuracy_score(y_test, predictions_test)
# TODO: Compute F-score on the the first 300 training samples
results['f_train'] = fbeta_score(y_train[:300], predictions_train, average='binary', beta=0.5)
# TODO: Compute F-score on the test set
results['f_test'] = fbeta_score(y_test, predictions_test, average='binary', beta=0.5)
# Success
print "{} trained on {} samples.".format(learner.__class__.__name__, sample_size)
# Return the results
return results
Implementation: Initial Model Evaluation¶
In the code cell, you will need to implement the following:
- Import the three supervised learning models you've discussed in the previous section.
- Initialize the three models and store them in
'clf_A','clf_B', and'clf_C'.- Use a
'random_state'for each model you use, if provided. - Note: Use the default settings for each model — you will tune one specific model in a later section.
- Use a
- Calculate the number of records equal to 1%, 10%, and 100% of the training data.
- Store those values in
'samples_1','samples_10', and'samples_100'respectively.
- Store those values in
Note: Depending on which algorithms you chose, the following implementation may take some time to run!
# TODO: Import the three supervised learning models from sklearn
from sklearn.ensemble import AdaBoostClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.svm import SVC
# TODO: Initialize the three models
clf_A = AdaBoostClassifier(random_state = 1)
clf_B = DecisionTreeClassifier(random_state = 1)
clf_C = SVC(random_state = 1)
# TODO: Calculate the number of samples for 1%, 10%, and 100% of the training data
samples_1 = len(X_train) / 100
samples_10 = len(X_train) / 10
samples_100 = len(X_train)
# Collect results on the learners
results = {}
for clf in [clf_A, clf_B, clf_C]:
clf_name = clf.__class__.__name__
results[clf_name] = {}
for i, samples in enumerate([samples_1, samples_10, samples_100]):
results[clf_name][i] = \
train_predict(clf, samples, X_train, y_train, X_test, y_test)
# Run metrics visualization for the three supervised learning models chosen
vs.evaluate(results, accuracy, fscore)
Improving Results¶
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 F-score.
Question 3 - Choosing the Best Model¶
Based on the evaluation you performed earlier, in one to two paragraphs, explain to CharityML which of the three models you believe to be most appropriate for the task of identifying individuals that make more than \$50,000.
Hint: Your answer should include discussion of the metrics, prediction/training time, and the algorithm's suitability for the data.
Answer:
Of the three models chosen, the best performing model on the task of identifying individudals who make more than $50,000 was the AdaBoost Classifier. It had a higher Accuracy Score and F-score on the Testing Set for each number of samples than either the Decision Trees Classifier or the Support Vector Classifier. One flaw with the Decision Tree Classifier is that it appears to have overfitted on the training data, as it performed much better on the 300 samples of training data than it did on the test data. One flaw with the Support Vector Classifier was the amount of time it took to train the classifier and then to make predictions, making it more challenging to iterate on and improve quickly (as we will do in later sections). It was also not able to produce an F-score on neither the training subset nor the testing subset - likely because it was unable to predict a positive class at all, which would make F1 equal to 0/0. For these reasons, and for the reason that the AdaBoost Classifier performs particularly well on questions of binary classification, it would appear to be the best model to choose for CharityML.
Question 4 - Describing the Model in Layman's Terms¶
In one to two paragraphs, explain to CharityML, 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:
The model we have chosen to use is the AdaBoost Classifier, which is an Ensemble Method technique. In simple terms, the AdaBoost Classifier is just a combination of a bunch of different learning algorithms that are slightly better at making predictions than flipping a coin. Based on how good they are at classifying the data, they have greater influence over the resulting model. One clever part of this technique is that its predictions are improved by simple models that classify the data in wholly different ways; the diversity in the classification of each model, the stronger the resulting weighted sum of the models is. A good way to think about the technique is to think about adding the models sequentially, and in a subsequent model placing less emphasis on points that had been correctly categorised previously and more emphasis on the points that were incorrectly categorised previously.
A very crude analogy one could make is mixing different types of alcohol on a night out. Let's pretend that getting drunk is like classifying data. If you drink solely beer, you will be able to get drunk, but not nearly as drunk as you would as if you were to also drink wine and tequila because of the different components of the alcohol (e.g. grain vs fruit vs cactus). Each alcohol has a different physiological effect on your body, and different sources interact in a way that heightens its effect. In this way, mixing different models together with differing weights can help improve classification. Interestingly, the more models included, the better the classification, in a similar fashion to the more different types of alcohol you drink, the drunker you become; this is a unique and somewhat counterintuitive property of boosting methods because one would assume adding more models would lead to overfitting, but it does not occur.
Implementation: Model Tuning¶
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:
- Import
sklearn.grid_search.GridSearchCVandsklearn.metrics.make_scorer. - Initialize the classifier you've chosen and store it in
clf.- Set a
random_stateif one is available to the same state you set before.
- Set a
- Create a dictionary of parameters you wish to tune for the chosen model.
- Example:
parameters = {'parameter' : [list of values]}. - Note: Avoid tuning the
max_featuresparameter of your learner if that parameter is available!
- Example:
- Use
make_scorerto create anfbeta_scorescoring object (with $\beta = 0.5$). - Perform grid search on the classifier
clfusing the'scorer', and store it ingrid_obj. - Fit the grid search object to the training data (
X_train,y_train), and store it ingrid_fit.
Note: Depending on the algorithm chosen and the parameter list, the following implementation may take some time to run!
# TODO: Import 'GridSearchCV', 'make_scorer', and any other necessary libraries
from sklearn.grid_search import GridSearchCV
from sklearn.metrics import make_scorer
from sklearn.ensemble import AdaBoostClassifier
# TODO: Initialize the classifier
clf = AdaBoostClassifier(random_state = 1)
# TODO: Create the parameters list you wish to tune
parameters = {'n_estimators' : [50,75,100,200], 'learning_rate' : [0.5,0.8,1,1.2]}
# TODO: Make an fbeta_score scoring object
scorer = make_scorer(fbeta_score, beta=0.5)
# TODO: Perform grid search on the classifier using 'scorer' as the scoring method
grid_obj = GridSearchCV(clf, parameters, scoring=scorer)
# TODO: Fit the grid search object to the training data and find the optimal parameters
grid_fit = grid_obj.fit(X_train, y_train)
# Get the estimator
best_clf = grid_fit.best_estimator_
# Make predictions using the unoptimized and model
predictions = (clf.fit(X_train, y_train)).predict(X_test)
best_predictions = best_clf.predict(X_test)
# Report the before-and-afterscores
print "Unoptimized model\n------"
print "Accuracy score on testing data: {:.4f}".format(accuracy_score(y_test, predictions))
print "F-score on testing data: {:.4f}".format(fbeta_score(y_test, predictions, beta = 0.5))
print "\nOptimized Model\n------"
print "Final accuracy score on the testing data: {:.4f}".format(accuracy_score(y_test, best_predictions))
print "Final F-score on the testing data: {:.4f}".format(fbeta_score(y_test, best_predictions, beta = 0.5))
display(pd.DataFrame(grid_obj.grid_scores_))
Question 5 - Final Model Evaluation¶
What is your optimized model's accuracy and F-score on the testing data? Are these scores better or worse than the unoptimized model? How do the results from your optimized model compare to the naive predictor benchmarks you found earlier in Question 1?
Note: Fill in the table below with your results, and then provide discussion in the Answer box.
Results:¶
| Metric | Benchmark Predictor | Unoptimized Model | Optimized Model |
|---|---|---|---|
| Accuracy Score | 0.2478 | 0.8576 | 0.8646 |
| F-score | 0.2917 | 0.7246 | 0.7365 |
Answer:
Both the unoptimized and optimized models performed much better than the naive predictor benchmarks of accuracy score and F-score that we found in Question 1. The optimized model also performed slightly better than the unoptimized model on both accuracy scores and F-scores.
Feature Importance¶
An important task when performing supervised learning on a dataset like the census data we study here is determining which features provide the most predictive power. By focusing on the relationship between only a few crucial features and the target label we simplify our understanding of the phenomenon, which is most always a useful thing to do. In the case of this project, that means we wish to identify a small number of features that most strongly predict whether an individual makes at most or more than \$50,000.
Choose a scikit-learn classifier (e.g., adaboost, random forests) that has a feature_importance_ attribute, which is a function that ranks the importance of features according to the chosen classifier. In the next python cell fit this classifier to training set and use this attribute to determine the top 5 most important features for the census dataset.
Question 6 - Feature Relevance Observation¶
When Exploring the Data, it was shown there are thirteen available features for each individual on record in the census data.
Of these thirteen records, which five features do you believe to be most important for prediction, and in what order would you rank them and why?
Answer:
- **Occupation:** There are certain occupations that typically earn workers more than others; in this dataset from a cursory glance it appeared that the categories 'Exec-managerial' and 'Prof-speciality' indicated higher incomes.
- **Education-num:** A typical econometrics model we would use as an example during my Economics degree was one that showed that having higher levels of education is correlated with higher income. One reason for this is a lot of well paying jobs typically require some sort of degree qualification.
- **Capital gain:** It would make sense that those who are earning higher incomes would also receive capital gains throughout the year, as they are more likely to have disposable income to allocate to their investment portfolios.
- **Hours worked:** My hunch is that hours worked around the 40 hour mark would have less of an explanatory impact, but those working say 15-20 hours a week would be more likely to be earning incomes lower than \$50k while those working 80 hours a week may be more likely to be earning incomes above \$50k.
- **Age:** Typically those who are older earn higher incomes than those who are younger because they have more years of experience under their belt.
Implementation - Extracting Feature Importance¶
Choose a scikit-learn supervised learning algorithm that has a feature_importance_ attribute available for it. This attribute is a function that ranks the importance of each feature when making predictions based on the chosen algorithm.
In the code cell below, you will need to implement the following:
- Import a supervised learning model from sklearn if it is different from the three used earlier.
- Train the supervised model on the entire training set.
- Extract the feature importances using
'.feature_importances_'.
# TODO: Import a supervised learning model that has 'feature_importances_'
from sklearn.ensemble import AdaBoostClassifier
# TODO: Train the supervised model on the training set
clf = AdaBoostClassifier(random_state = 1)
model = clf.fit(X_train, y_train)
# TODO: Extract the feature importances
importances = model.feature_importances_
# Plot
vs.feature_plot(importances, X_train, y_train)
# look at top "n" feature importances
n = 10
fi = model.feature_importances_
pd.Series(fi, index=X_train.columns).sort_values()[-n:].plot(kind='barh');
Question 7 - Extracting Feature Importance¶
Observe the visualization created above which displays the five most relevant features for predicting if an individual makes at most or above \$50,000.
How do these five features compare to the five features you discussed in Question 6? If you were close to the same answer, how does this visualization confirm your thoughts? If you were not close, why do you think these features are more relevant?
Answer:
I was pretty close! I did not predict capital-loss being as strong a feature as capital-gain, and my #1 choice of Occupation didn't make the top 5, though all the rest made it (albeit in a different order). I think it makes sense that age, capital gain/loss, hours-per-week, and education-num would have a strong effect as I outlined above, though I would have expected education-num to have had a bigger effect than age. I do also have a hunch that the features importances function has a preference for numerical rather than categorical data.
Feature Selection¶
How does a model perform if we only use a subset of all the available features in the data? With less features required to train, the expectation is that training and prediction time is much lower — at the cost of performance metrics. From the visualization above, we see that the top five most important features contribute more than half of the importance of all features present in the data. This hints that we can attempt to reduce the feature space and simplify the information required for the model to learn. The code cell below will use the same optimized model you found earlier, and train it on the same training set with only the top five important features.
# Import functionality for cloning a model
from sklearn.base import clone
# Reduce the feature space
X_train_reduced = X_train[X_train.columns.values[(np.argsort(importances)[::-1])[:5]]]
X_test_reduced = X_test[X_test.columns.values[(np.argsort(importances)[::-1])[:5]]]
# Train on the "best" model found from grid search earlier
clf = (clone(best_clf)).fit(X_train_reduced, y_train)
# Make new predictions
reduced_predictions = clf.predict(X_test_reduced)
# Report scores from the final model using both versions of data
print "Final Model trained on full data\n------"
print "Accuracy on testing data: {:.4f}".format(accuracy_score(y_test, best_predictions))
print "F-score on testing data: {:.4f}".format(fbeta_score(y_test, best_predictions, beta = 0.5))
print "\nFinal Model trained on reduced data\n------"
print "Accuracy on testing data: {:.4f}".format(accuracy_score(y_test, reduced_predictions))
print "F-score on testing data: {:.4f}".format(fbeta_score(y_test, reduced_predictions, beta = 0.5))
Question 8 - Effects of Feature Selection¶
How does the final model's F-score and accuracy score on the reduced data using only five features compare to those same scores when all features are used?
If training time was a factor, would you consider using the reduced data as your training set?
Answer:
The final model trained on full data outperformed the final model trained on reduced data, with an accuracy score of 0.8646 > 0.8357 and F-score of 0.7365 > 0.6850. If the dataset was sufficiently large or that the chosen model took a particularly long time to train (e.g. an Support Vector Classifier) that training time was an important factor to consider, I think it would be fine to use the final model trained on reduced data, though it would not be ideal.
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.