Notebook

Finding Donors for CharityML¶

CharityML is a fictitious charity organization located in the heart of Silicon Valley that was established to provide financial support for people eager to learn machine learning. After nearly 32,000 letters were sent to people in the community, CharityML determined that every donation they received came from someone that was making more than $50,000 annually. To expand their potential donor base, CharityML has decided to send letters to residents of California, but to only those most likely to donate to the charity. With nearly 15 million working Californians, CharityML has brought you on board to help build an algorithm to best identify potential donors and reduce overhead cost of sending mail. Your goal will be evaluate and optimize several different supervised learners to determine which algorithm will provide the highest donation yield while also reducing the total number of letters being sent.

Credits¶

This notebook is a part of my learning path based on Data Scientist Nanodegree Program enrollment.

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.

In [2]:

# 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 scripts.visuals as vs

# Pretty display for notebooks
%matplotlib inline

# Load the Census dataset
data = pd.read_csv("../../data/census.csv")

# Success - Display the first record
display(data.head(n=1))

	age	workclass	education_level	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	39	State-gov	Bachelors	13.0	Never-married	Adm-clerical	Not-in-family	White	Male	2174.0	0.0	40.0	United-States	<=50K

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.

In [2]:

# TODO: Total number of records
n_records = data.shape[0]

# TODO: Number of records where individual's income is more than $50,000
n_greater_50k = data[data["income"] == ">50K"].shape[0]

# TODO: Number of records where individual's income is at most $50,000
n_at_most_50k = data[data["income"] == "<=50K"].shape[0]

# TODO: Percentage of individuals whose income is more than $50,000
greater_percent = n_greater_50k/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: {}%".format(greater_percent))

Total number of records: 45222
Individuals making more than $50,000: 11208
Individuals making at most $50,000: 34014
Percentage of individuals making more than $50,000: 24.78439697492371%

It is clear that the 2 classes (individuals with income > $50k = 11208 and individuals with income atmost $50k = 34014) are imbalanced. Please check this link to understand how to deal with imbalanced data.

** Featureset Exploration **

age: continuous.
workclass: Private, Self-emp-not-inc, Self-emp-inc, Federal-gov, Local-gov, State-gov, Without-pay, Never-worked.
education: Bachelors, Some-college, 11th, HS-grad, Prof-school, Assoc-acdm, Assoc-voc, 9th, 7th-8th, 12th, Masters, 1st-4th, 10th, Doctorate, 5th-6th, Preschool.
education-num: continuous.
marital-status: Married-civ-spouse, Divorced, Never-married, Separated, Widowed, Married-spouse-absent, Married-AF-spouse.
occupation: Tech-support, Craft-repair, Other-service, Sales, Exec-managerial, Prof-specialty, Handlers-cleaners, Machine-op-inspct, Adm-clerical, Farming-fishing, Transport-moving, Priv-house-serv, Protective-serv, Armed-Forces.
relationship: Wife, Own-child, Husband, Not-in-family, Other-relative, Unmarried.
race: Black, White, Asian-Pac-Islander, Amer-Indian-Eskimo, Other.
sex: Female, Male.
capital-gain: continuous.
capital-loss: continuous.
hours-per-week: continuous.
native-country: United-States, Cambodia, England, Puerto-Rico, Canada, Germany, Outlying-US(Guam-USVI-etc), India, Japan, Greece, South, China, Cuba, Iran, Honduras, Philippines, Italy, Poland, Jamaica, Vietnam, Mexico, Portugal, Ireland, France, Dominican-Republic, Laos, Ecuador, Taiwan, Haiti, Columbia, Hungary, Guatemala, Nicaragua, Scotland, Thailand, Yugoslavia, El-Salvador, Trinadad&Tobago, Peru, Hong, Holand-Netherlands.

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.

In [3]:

# 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.

In [4]:

# Log-transform the skewed features
skewed = ['capital-gain', 'capital-loss']
features_log_transformed = pd.DataFrame(data = features_raw)
features_log_transformed[skewed] = features_raw[skewed].apply(lambda x: np.log(x + 1))

# Visualize the new log distributions
vs.distribution(features_log_transformed, 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.

In [5]:

# Import sklearn.preprocessing.StandardScaler
from sklearn.preprocessing import MinMaxScaler

# Initialize a scaler, then apply it to the features
scaler = MinMaxScaler() # default=(0, 1)
numerical = ['age', 'education-num', 'capital-gain', 'capital-loss', 'hours-per-week']

features_log_minmax_transform = pd.DataFrame(data = features_log_transformed)
features_log_minmax_transform[numerical] = scaler.fit_transform(features_log_transformed[numerical])

# Show an example of a record with scaling applied
display(features_log_minmax_transform.head(n = 5))

	age	workclass	education_level	education-num	marital-status	occupation	relationship	race	sex	capital-gain	hours-per-week	native-country
0	0.301370	State-gov	Bachelors	0.800000	Never-married	Adm-clerical	Not-in-family	White	Male	0.667492	0.397959	United-States
1	0.452055	Self-emp-not-inc	Bachelors	0.800000	Married-civ-spouse	Exec-managerial	Husband	White	Male	0.000000	0.122449	United-States
2	0.287671	Private	HS-grad	0.533333	Divorced	Handlers-cleaners	Not-in-family	White	Male	0.000000	0.397959	United-States
3	0.493151	Private	11th	0.400000	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0.000000	0.397959	United-States
4	0.150685	Private	Bachelors	0.800000	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0.000000	0.397959	Cuba

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_log_minmax_transform' data.
Convert the target label 'income_raw' to numerical entries.
- Set records with "<=50K" to 0 and records with ">50K" to 1.

In [6]:

# TODO: One-hot encode the 'features_log_minmax_transform' data using pandas.get_dummies()
features_final = pd.get_dummies(features_log_minmax_transform)

# TODO: Encode the 'income_raw' data to numerical values
income = (income_raw == ">50K") * 1

# Print the number of features after one-hot encoding
encoded = list(features_final.columns)
print("{} total features after one-hot encoding.".format(len(encoded)))

# Uncomment the following line to see the encoded feature names
# print(encoded)

103 total features after one-hot encoding.

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.

In [7]:

# 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_final, 
                                                    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]))

Training set has 36177 samples.
Testing set has 9045 samples.

/opt/conda/lib/python3.6/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)

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.

Note: Recap of accuracy, precision, recall¶

** Accuracy ** measures how often the classifier makes the correct prediction. It’s the ratio of the number of correct predictions to the total number of predictions (the number of test data points).

** Precision ** tells us what proportion of messages we classified as spam, actually were spam. It is a ratio of true positives(words classified as spam, and which are actually spam) to all positives(all words classified as spam, irrespective of whether that was the correct classificatio), in other words it is the ratio of

[True Positives/(True Positives + False Positives)]

** Recall(sensitivity)** tells us what proportion of messages that actually were spam were classified by us as spam. It is a ratio of true positives(words classified as spam, and which are actually spam) to all the words that were actually spam, in other words it is the ratio of

[True Positives/(True Positives + False Negatives)]

For classification problems that are skewed in their classification distributions like in our case, for example if we had a 100 text messages and only 2 were spam and the rest 98 weren't, accuracy by itself is not a very good metric. We could classify 90 messages as not spam(including the 2 that were spam but we classify them as not spam, hence they would be false negatives) and 10 as spam(all 10 false positives) and still get a reasonably good accuracy score. For such cases, precision and recall come in very handy. These two metrics can be combined to get the F1 score, which is weighted average(harmonic mean) of the precision and recall scores. This score can range from 0 to 1, with 1 being the best possible F1 score(we take the harmonic mean as we are dealing with ratios).

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? You must use the code cell below and assign your results to 'accuracy' and 'fscore' to be used later.

** Please note ** that the the purpose of generating a naive predictor is simply to show what a base model without any intelligence would look like. In the real world, ideally your base model would be either the results of a previous model or could be based on a research paper upon which you are looking to improve. When there is no benchmark model set, getting a result better than random choice is a place you could start from.

** HINT: **

When we have a model that always predicts '1' (i.e. the individual makes more than 50k) then our model will have no True Negatives(TN) or False Negatives(FN) as we are not making any negative('0' value) predictions. Therefore our Accuracy in this case becomes the same as our Precision(True Positives/(True Positives + False Positives)) as every prediction that we have made with value '1' that should have '0' becomes a False Positive; therefore our denominator in this case is the total number of records we have in total.
Our Recall score(True Positives/(True Positives + False Negatives)) in this setting becomes 1 as we have no False Negatives.

In [8]:

'''
TP = np.sum(income) # Counting the ones as this is the naive case. Note that 'income' is the 'income_raw' data 
encoded to numerical values done in the data preprocessing step.
FP = income.count() - TP # Specific to the naive case

TN = 0 # No predicted negatives in the naive case
FN = 0 # No predicted negatives in the naive case
'''

# TODO: Calculate accuracy, precision and recall
TP = np.sum(income)
FP = income.count() - TP
TN = 0
FN = 0

accuracy = (TP + TN)/(TP + FP + TN + FN)
precision = TP/(TP + FP)
recall = TP/(TP + FN)

# TODO: Calculate F-score using the formula above for beta = 0.5 and correct values for precision and recall.
beta = 0.5
fscore = (1 + beta ** 2) * precision * recall/(beta ** 2 * precision + recall)

# Print the results 
print("Naive Predictor: [Accuracy score: {:.4f}, F-score: {:.4f}]".format(accuracy, fscore))

Naive Predictor: [Accuracy score: 0.2478, F-score: 0.2917]

Note that as TN = 0 and FN = 0; both the accuracy and precission in our case are the same. To understand more about the right metrics for classification problems please look at Beyond Accuracy: Precision and Recall.

Supervised Learning Models¶

The following are some of the supervised learning models that 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.
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?

** HINT: **

Structure your answer in the same format as above^, with 4 parts for each of the three models you pick. Please include references with your answer.

Answer:

We are dealing with classification problem. Based on [1] let's choose the following models:

Decision Trees
Ensemble Methods (AdaBoost)
Support Vector Machines (SVM)

Decision Trees¶

Describe one real-world application in industry where the model can be applied.

Predicting stock markets movement.

What are the strengths of the model; when does it perform well?

Easy to interpret and explain; fairly robust to outliers; can learn non-linear relationships.

What are the weaknesses of the model; when does it perform poorly?

Prone to overfiting; sensitive to small pertubations in the data.

What makes this model a good candidate for the problem, given what you know about the data?

Decision tree models are widely used in the binay classification but care should be taken where there is a lot of features.

Ensemble Methods (AdaBoost)¶

Describe one real-world application in industry where the model can be applied.

Areas of video and image recognition, for example traffic sign recognition [2]

What are the strengths of the model; when does it perform well?

Improve generalization using multiple weak classifiers; We can combine simple models to build a complex one.

What are the weaknesses of the model; when does it perform poorly?

Sensitive to noise and outliers; compute complex and intensive;

What makes this model a good candidate for the problem, given what you know about the data?

AdaBoost ensemble method can help us analyze the relationship between the various features and their effect on the probability of the income thus creating better model.

Support Vector Machines (SVM)¶

Describe one real-world application in industry where the model can be applied.

Face detection [3].

What are the strengths of the model; when does it perform well?

With the kernel trick it can find optimal boundaries; performs well when there is clean margin of data separation; good for smaller dataset; plays nicely with highly dimensional data.

What are the weaknesses of the model; when does it perform poorly?

To much noise in the data can lead to overfit to them; computationally intensive; performs poorly on large datasets.

What makes this model a good candidate for the problem, given what you know about the data?

There are a lot of features in our data (mostly because of our one-hot encoding feature transformation) where the SVM should help.

[1] Modern Machine Learning Algorithms: Strengths and Weaknesses
[2] Traffic Sign Recognition using Evolutionary Adaboost detection and Forest-ECOC classification
[3] Real-Life Applications of SVM (Support Vector Machines) [4] Machine Learning Tutorials

Knowing the pros and the cons of each model will greatly help to decide which model to select for the given problem/dataset. Please take a look on this cheat sheet to understand more about model selection.

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_score and accuracy_score from sklearn.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 points X_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 beta parameter!

In [9]:

# TODO: Import two metrics from sklearn - fbeta_score and accuracy_score
from sklearn.metrics import fbeta_score, 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' using .fit(training_features[:], training_labels[:])
    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(X_test),
    #       then get predictions on the first 300 training samples(X_train) using .predict()
    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 which is y_train[:300]
    results['acc_train'] = accuracy_score(y_train[:300], predictions_train)
        
    # TODO: Compute accuracy on test set using accuracy_score()
    results['acc_test'] = accuracy_score(y_test, predictions_test)
    
    # TODO: Compute F-score on the the first 300 training samples using fbeta_score()
    results['f_train'] = fbeta_score(y_train[:300], predictions_train, average = 'binary', beta = 0.5)
        
    # TODO: Compute F-score on the test set which is y_test
    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

You can learn more about the use of pipelines in ML reading the scikit-learn documentation.

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.
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.

Note: Depending on which algorithms you chose, the following implementation may take some time to run!

In [10]:

# TODO: Import the three supervised learning models from sklearn
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import AdaBoostClassifier
from sklearn.svm import SVC

# For more information why to use random state, please take a look on:
# https://stackoverflow.com/questions/45089858/what-does-the-random-state-parameter-do-in-sklearns-parametersampler/45089955#45089955
random_state = 2018

# TODO: Initialize the three models
clf_A = DecisionTreeClassifier(random_state = random_state)
clf_B = AdaBoostClassifier(random_state = random_state)
clf_C = SVC(random_state = random_state)

# TODO: Calculate the number of samples for 1%, 10%, and 100% of the training data
# HINT: samples_100 is the entire training set i.e. len(y_train)
# HINT: samples_10 is 10% of samples_100 (ensure to set the count of the values to be `int` and not `float`)
# HINT: samples_1 is 1% of samples_100 (ensure to set the count of the values to be `int` and not `float`)
samples_100 = len(X_train)
samples_10 = int(len(X_train)/10)
samples_1 = int(len(X_train)/100)

# 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)

DecisionTreeClassifier trained on 361 samples.
DecisionTreeClassifier trained on 3617 samples.
DecisionTreeClassifier trained on 36177 samples.
AdaBoostClassifier trained on 361 samples.
AdaBoostClassifier trained on 3617 samples.
AdaBoostClassifier trained on 36177 samples.

/opt/conda/lib/python3.6/site-packages/sklearn/metrics/classification.py:1135: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no predicted samples.
  'precision', 'predicted', average, warn_for)

SVC trained on 361 samples.
SVC trained on 3617 samples.
SVC trained on 36177 samples.

In [11]:

# Printing out the values as the chart for time is hard to read for AdaBoostClassifier and DecissionTreeClassifier

print("DecisionTreeClassifier:\t [Prediction time: {:.4f}, Train time: {:.4f}]".format(results['DecisionTreeClassifier'][2]['pred_time'], results['DecisionTreeClassifier'][2]['train_time']))
print("AdaBoostClassifier:\t [Prediction time: {:.4f}, Train time: {:.4f}]".format(results['AdaBoostClassifier'][2]['pred_time'], results['AdaBoostClassifier'][2]['train_time']))
print("SVC:\t\t\t [Prediction time: {:.4f}, Train time: {:.4f}]".format(results['SVC'][2]['pred_time'], results['SVC'][2]['train_time']))

DecisionTreeClassifier:	 [Prediction time: 0.0058, Train time: 0.4156]
AdaBoostClassifier:	 [Prediction time: 0.0934, Train time: 1.5297]
SVC:			 [Prediction time: 16.6950, Train time: 102.7854]

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: ** Look at the graph at the bottom left from the cell above(the visualization created by vs.evaluate(results, accuracy, fscore)) and check the F score for the testing set when 100% of the training set is used. Which model has the highest score? Your answer should include discussion of the:

metrics - F score on the testing when 100% of the training data is used,
prediction/training time
the algorithm's suitability for the data.

Answer:

The best performing model for identifying a person who make more than $50,000 is AdaBoost Classifier. Looking at the accuracy score and F-score of the testing set we can clearly see that AdaBoost Classifier outperforms the Decision Tree Classifier and SVC. It has the highest F-score value on 100% training test size.
The performance metrics for the training set show us that the Decision Tree Classifier overfit on the training data (accuracy and F-score near 1).
The time needed to train SVC model is enormous comapred to other models. It is a bottleneck to improve quickly and scale. It's also not able to produce an F-score on neither the training subset nor the testing subset. The reason is that the model is unabled to predict a positive class at all. For the AdaBoost model the training as well as prediction time looks reasonable (on the level of 1 second for the training time and mili seconds for the prediction time).
With high accuracy and F-score, good timing, AdaBoost Classifier looks the most promissing supervised learning model for the CharityML problem.

In the recent years one algorithm emerged as favourite in the machine learning community, it is actually one of the most used in Kaggle: Xgboost. Here you can find an informative discussion on why that is the case.
The algorithm is not available sci-kit learn, here is how you can start working with it.

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 jargon, such as describing equations.

** HINT: **

When explaining your model, if using external resources please include all citations.

Answer:

AdaBoost is an ensemble technique that attempts to create a strong classifier from a number of weak classifiers. A weak classifier is a classifier that simply does better than random guessing but still not necessarily that well overall. The idea is to set weights to both classifiers and data points (samples) in a way that forces classifiers to concentrate on observations that are difficult to correctly classify. This process is done sequentially in that the two weights are adjusted at each step as iterations of the algorithm proceed. This is why Adaboost is referred to as a sequential ensemble method — ensemble referring to a type of learning that combines several models to improve the final predictive performance

What is A Good Weak Learner?
A weak learner is any machine learning algorithm that gives better accuracy than simply guessing. For instance, if you are trying to classify animals at a zoo, you might have an algorithm that can correctly identify zebras most of the time, but it simply guesses for any other animal. That algorithm would be a weak learner because it is better than guessing.
If you had an algorithm that identified every animal as a zebra, then that probably is not better than guessing and so it would not be a weak learner.
For boosting problems, the best kinds of weak learners are ones that are very accurate, even if it is only over a limited scope of the problem. For instance the algorithm that correctly identifies zebras would be good. It allows you to confidently identify as least most of the zebras, allowing other weak learners to focus on the remaining animals.

How Are Weak Learners Combined?
Boosting algorithms typically work by solving subsections of the problem, by peeling them away so future boosting iterations can solve the remaining sections.
Imagine you are hiring people to build your house, and you have 10 different big jobs that need to be done. A great way of doing it would be to get someone who is really good at foundations to build the foundation. Then hire a very good carpenter to focus on the framing. Then hire a great roofer and plumber to focus stage, a small subsection of the project is getting completely solved.
The takeaway is that weak learners are best combined in a way that allows each one to solve a limited section of the problem. Any machine learning routine can be used as a weak learner. Neural nets, support vector machines or any other would work, but the most commonly used weak learner is the decision tree.

References:

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.GridSearchCV and sklearn.metrics.make_scorer.
Initialize the classifier you've chosen and store it in clf.
Set a random_state if one is available to the same state you set before.
Create a dictionary of parameters you wish to tune for the chosen model.
Example: parameters = {'parameter' : [list of values]}.
Note: Avoid tuning the max_features parameter of your learner if that parameter is available!
Use make_scorer to create an fbeta_score scoring object (with $\beta = 0.5$).
Perform grid search on the classifier clf using the 'scorer', and store it in grid_obj.
Fit the grid search object to the training data (X_train, y_train), and store it in grid_fit.

Note: Depending on the algorithm chosen and the parameter list, the following implementation may take some time to run!

In [12]:

# TODO: Import 'GridSearchCV', 'make_scorer', and any other necessary libraries
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import make_scorer

# TODO: Initialize the classifier
clf = AdaBoostClassifier(random_state = random_state)

# TODO: Create the parameters list you wish to tune, using a dictionary if needed.
# HINT: parameters = {'parameter_1': [value1, value2], 'parameter_2': [value1, value2]}
parameters = {
    'n_estimators': [50, 100, 150, 200],
    'learning_rate': [0.1, 0.5, 1.0, 1.5]
}

# TODO: Make an fbeta_score scoring object using make_scorer()
scorer = make_scorer(fbeta_score, beta = 0.5)

# TODO: Perform grid search on the classifier using 'scorer' as the scoring method using GridSearchCV()
grid_obj = GridSearchCV(clf, parameters, scorer)

# TODO: Fit the grid search object to the training data and find the optimal parameters using fit()
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)))

Unoptimized model
------
Accuracy score on testing data: 0.8576
F-score on testing data: 0.7246

Optimized Model
------
Final accuracy score on the testing data: 0.8645
Final F-score on the testing data: 0.7375

In [13]:

import seaborn as sns

gridResults = grid_fit.grid_scores_
gridResultsDf = pd.DataFrame([[r[0]['n_estimators'],r[0]['learning_rate'],r[1]] for r in gridResults],
                             columns = ['n_estimators','learning_rate','score'])

sns.heatmap(gridResultsDf.pivot(columns='n_estimators',index='learning_rate', values='score'), annot=True);

/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_search.py:761: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20
  DeprecationWarning)

We could actually go well beyond grid search and implement pipelines where the whole machine learning process becomes 'grid-searchable' and you can parameterize and search the whole process though cross validation. We can try out several algorithms automatically as well too! Watch out though this is pretty advanced stuff, here is a great, informative, top notch tutorial from Zac Sewart!

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	Unoptimized Model	Optimized Model
Accuracy Score	0.8576	0.8645
F-score	0.7246	0.7375

Answer:

The optimized model has an accuracy of 0.8645 and an F-score of 0.7375 which is better than the unoptimized model. Comparing the results with the Naive Predictor from Question 1 (accuracy score: 0.2478, F-score: 0.2917) it is a huge improvement.

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:

Trying to establish the maximum amount of income, I would choose the following 5 features:

*occupation* - some occupations give the possibility to earn more than the others.
*education-num* - a well paying jobs typically require some sort of degree qualification.
*age* - we can assume that the older you are the more you earn because you have more years of experience but at some point an individual could be retired and earn less.
*hours-per-week* - working hours per week likely influence the income; I assume that part-time or unemployed people will almost always eran less.
*capital-gain* - people with more capital gain are more likely to earn more than 50K; it indicates that they have enough money to do some investements.

Feature selection is a very critical step in any Machine Learning algorithm's workflow. Top reasons to use feature selection are:

it enables the algorithm to train faster,
it reduce the complexity of a model and makes it easier to interpret,
it improves the accuracy of a model if the right subset is choosen.

Implementation - Extracting Feature Importance¶

Choose a scikit-learn supervised learning algorithm that has a feature_importance_ attribute availble 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_'.

In [14]:

# TODO: Import a supervised learning model that has 'feature_importances_'
from sklearn.ensemble import AdaBoostClassifier

# TODO: Train the supervised model on the training set using .fit(X_train, y_train)
clf = AdaBoostClassifier(random_state = random_state)
model = clf.fit(X_train, y_train)

# TODO: Extract the feature importances using .feature_importances_ 
importances = model.feature_importances_

# Plot
vs.feature_plot(importances, X_train, y_train)

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:

4 out of 5 features were correctly identified as relevant for selecting individual making at most $50,000 or above but not in the correct order. I'm a little bit confused that both capital-gain and capital-loss are included as high ranked. I thought that both features are dependant and for me capital-gain was more valuable. I assumed capital-loss suggests a loss in earnings but it appears to be another indicator of handling large investments and therefore better income. Feature like occupation is categorical one and it looks like the model behaves better with numerical features.

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.

In [15]:

# 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)))

Final Model trained on full data
------
Accuracy on testing data: 0.8645
F-score on testing data: 0.7375

Final Model trained on reduced data
------
Accuracy on testing data: 0.8379
F-score on testing data: 0.6898

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 reduced model performes worse than the optimized model with all features enabled. As such it looks like using reduced data configuration is subject to discussion but it is still better than the unoptimized model.
I would consider using a reduced dataset if training time was a factor depending on how important those extra percentage points of accuracy and F-score are.