Bootstrap aggregation aka bagging is a technique used in machine learning that relies on resampling from the sample and running multiple models from the different samples. The mean or some other value is calculated from the results of each model. For example, if you are using Decisions trees, bagging would have you run the model several times with several different subsamples to help deal with variance in statistics.

Bagging is an excellent tool for algorithms that are considered weaker or more susceptible to variances such as decision trees or KNN. In this post, we will use bagging to develop a model that determines whether or not people voted using the turnout dataset. These results will then be compared to a model that was developed in a traditional way.

We will use the turnout dataset available in the pydataset module. Below is some initial code.

from pydataset import data
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import BaggingClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import cross_val_score
from sklearn.metrics import classification_report

We will load our dataset. Then we will separate the independnet and dependent variables from each other and create our train and test sets. The code is below.

df=data("turnout")
X=df[['age','educate','income',]]
y=df['vote']
X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=.3,random_state=0)

We can now prepare to run our model. We need to first set up the bagging function. There are several arguments that need to be set. The max_samples argument determines the largest amount of the dataset to use in resampling. The max_features argument is the max number of features to use in a sample. Lastly, the n_estimators is for determining the number of subsamples to draw. The code is as follows

 h=BaggingClassifier(KNeighborsClassifier(n_neighbors=7),max_samples=0.7,max_features=0.7,n_estimators=1000)

Basically, what we told python was to use up to 70% of the samples, 70% of the features, and make 100 different KNN models that use seven neighbors to classify. Now we run the model with the fit function, make a prediction with the predict function, and check the accuracy with the classificarion_reoirts function.

h.fit(X_train,y_train)
y_pred=h.predict(X_test)
print(classification_report(y_test,y_pred))

This looks oka below are the results when you do a traditional model without bagging

X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=.3,random_state=0)
clf=KNeighborsClassifier(7)
clf.fit(X_train,y_train)
y_pred=clf.predict(X_test)
print(classification_report(y_test,y_pred))

The improvement is not much. However, this depends on the purpose and scale of your project. A small improvement can mean millions in the reight context such as for large companies such as Google who deal with billions of people per day.

Conclusion

This post provides an example of the use of bagging in the context of classification. Bagging provides a why to improve your model through the use of resampling.

Support vector machines (SVM) is an algorithm used to fit non-linear models. The details are complex but to put it simply  SVM tries to create the largest boundaries possible between the various groups it identifies in the sample. The mathematics behind this is complex especially if you are unaware of what a vector is as defined in algebra.

This post will provide an example of SVM using Python broken into the following steps.

1. Data preparation
2. Model Development

We will use two different kernels in our analysis. The linear kernel and he rbf kernel. The difference in terms of kernels has to do with how the boundaries between the different groups are made.

Data Preparation

We are going to use the OFP dataset available in the pydataset module. We want to predict if someone single or not. Below is some initial code.

import numpy as np
import pandas as pd
from pydataset import data
from sklearn import svm
from sklearn.metrics import classification_report
from sklearn import model_selection

We now need to load our dataset and remove any missing values.

df=pd.DataFrame(data('OFP'))
df=df.dropna()
df.head()

Looking at the dataset we need to do something with the variables that have text. We will create dummy variables for all except region and hlth. The code is below.

dummy=pd.get_dummies(df['black'])
df=pd.concat([df,dummy],axis=1)
df=df.rename(index=str, columns={"yes": "black_person"})
df=df.drop('no', axis=1)

dummy=pd.get_dummies(df['sex'])
df=pd.concat([df,dummy],axis=1)
df=df.rename(index=str, columns={"male": "Male"})
df=df.drop('female', axis=1)

dummy=pd.get_dummies(df['employed'])
df=pd.concat([df,dummy],axis=1)
df=df.rename(index=str, columns={"yes": "job"})
df=df.drop('no', axis=1)

dummy=pd.get_dummies(df['maried'])
df=pd.concat([df,dummy],axis=1)
df=df.rename(index=str, columns={"no": "single"})
df=df.drop('yes', axis=1)

dummy=pd.get_dummies(df['privins'])
df=pd.concat([df,dummy],axis=1)
df=df.rename(index=str, columns={"yes": "insured"})
df=df.drop('no', axis=1)

For each variable, we did the following

1. Created a dummy in the dummy dataset
2. Combined the dummy variable with our df dataset
3. Renamed the dummy variable based on yes or no
4. Drop the other dummy variable from the dataset. Python creates two dummies instead of one.

If you look at the dataset now you will see a lot of variables that are not necessary. Below is the code to remove the information we do not need.

df=df.drop(['black','sex','maried','employed','privins','medicaid','region','hlth'],axis=1)
df.head()

This is much cleaner. Now we need to scale the data. This is because SVM is sensitive to scale. The code for doing this is below.

df = (df - df.min()) / (df.max() - df.min())
df.head()

We can now create our dataset with the independent variables and a separate dataset with our dependent variable. The code is as follows.

X=df[['ofp','ofnp','opp','opnp','emr','hosp','numchron','adldiff','age','school','faminc','black_person','Male','job','insured']]
y=df['single']

We can now move to model development

Model Development

We need to make our test and train sets first. We will use a 70/30 split.

X_train,X_test,y_train,y_test=model_selection.train_test_split(X,y,test_size=.3,random_state=1)

Now, we need to create the models or the hypothesis we want to test. We will create two hypotheses. The first model is using a linear kernel and the second is one using the rbf kernel. For each of these kernels, there are hyperparameters that need to be set which you will see in the code below.

h1=svm.LinearSVC(C=1)
h2=svm.SVC(kernel='rbf',degree=3,gamma=0.001,C=1.0)

The details about the hyperparameters are beyond the scope of this post. Below are the results for the first model.

The overall accuracy is 73%. The crosstab() function provides a breakdown of the results and the classification_report() function provides other metrics related to classification. In this situation, 0 means not single or married while 1 means single. Below are the results for model 2

You can see the results are similar with the first model having a slight edge. The second model really struggls with predicting people who are actually single. You can see thtat the recall in particular is really poor.

Conclusion

This post provided how to ob using SVM in python. How this algorithm works can be somewhat confusing. However, its use can be powerful if use appropriately.

In this post, we will learn to do some basic exploration of a dataframe in Python. Some of the task we will complete include the following…

• Import data
• Examine data
• Work with strings
• Calculating descriptive statistics

Import Data 

First, you need data, therefore, we will use the Titanic dataset, which is readily available on the internet. We will need to use the pd.read_csv() function from the pandas package. This means that we must also import pandas. Below is the code.

import pandas as pd
df=pd.read_csv('FILE LOCATION HERE')

In the code above we imported pandas as pd so we can use the functions within it. Next, we create an object called ‘df’. Inside this object, we used the pd.read_csv() function to read our file into the system. The location of the file needs to type in quotes inside the parentheses. Having completed this we can now examine the data.

Data Examination

Now we want to get an idea of the size of our dataset, any problems with missing. To determine the size we use the .shape function as shown below.

df.shape
Out[33]: (891, 12)

Results indicate that we have 891 rows and 12 columns/variables. You can view the whole dataset by typing the name of the dataframe “df” and pressing enter. If you do this you may notice there are a lot of NaN values in the “Cabin” variable. To determine exactly how many we can use is.null() in combination with the values_count. variables.

df['Cabin'].isnull().value_counts()
Out[36]: 
True     687
False    204
Name: Cabin, dtype: int64

The code starts with the name of the dataframe. In the brackets, you put the name of the variable. After that, you put the functions you are using. Keep in mind that the order of the functions matters. You can see we have over 200 missing examples. For categorical varable, you can also see how many examples are part of each category as shown below.

df['Embarked'].value_counts()
Out[39]: 
S    644
C    168
Q     77
Name: Embarked, dtype: int64

This time we used our ‘Embarked’ variable. However, we need to address missing values before we can continue. To deal with this we will use the .dropna() function on the dataset. THen we will check the size of the dataframe again with the “shape” function.

df=df.dropna(how='any')
df.shape
Out[40]: (183, 12)

You can see our dataframe is much smaller going 891 examples to 183. We can now move to other operations such as dealing with strings.

Working with Strings

What you do with strings really depends or your goals. We are going to look at extraction, subsetting, determining the length. Our first step will be to extract the last name of the first five people. We will do this with the code below.

df['Name'][0:5].str.extract('(\w+)')
Out[44]: 
1 Cumings
3 Futrelle
6 McCarthy
10 Sandstrom
11 Bonnell
Name: Name, dtype: object

As you can see we got the last names of the first five examples. We did this by using the following format…

dataframe name[‘Variable Name’].function.function(‘whole word’))

.str is a function for dealing with strings in dataframes. The .extract() function does what its name implies.

If you want, you can even determine how many letters each name is. We will do this with the .str and .len() function on the first five names in the dataframe.

df['Name'][0:5].str.len()
Out[64]: 
1 51
3 44
6 23
10 31
11 24
Name: Name, dtype: int64

Hopefully, the code is becoming easier to read and understand.

Aggregation

We can also calculate some descriptive statistics. We will do this for the “Fare” variable. The code is repetitive in that only the function changes so we will run all of them at once. Below we are calculating the mean, max, minimum, and standard deviation  for the price of a fare on the Titanic

df['Fare'].mean()
Out[77]: 78.68246885245901

df['Fare'].max()
Out[78]: 512.32920000000001

df['Fare'].min()
Out[79]: 0.0

df['Fare'].std()
Out[80]: 76.34784270040574

Conclusion

This post provided you with some ways in which you can maneuver around a dataframe in Python.

In this post, we are going to explore arrays is created by the numpy package in Python. Understanding how arrays are created and manipulated is useful when you need to perform complex coding and or analysis. In particular, we will address the following,

1. Creating and exploring arrays
2. Math with arrays
3. Manipulating arrays

Creating and Exploring an Array

Creating an array is simple. You need to import the numpy package and then use the np.array function to create the array. Below is the code.

import numpy as np
example=np.array([[1,2,3,4,5],[6,7,8,9,10]])

Making an array requires the use of square brackets. If you want multiple dimensions or columns than you must use inner square brackets. In the example above I made an array with two dimensions and each dimension has it’s own set of brackets.

Also, notice that we imported numpy as np. This is a shorthand so that we do not have to type the word numpy but only np. In addition, we now created an array with ten data points spread in two dimensions.

There are several functions you can use to get an idea of the size of a data set. Below is a list with the function and explanation.

• .ndim = number of dimensions
• .shape =  Shares the number of rows and columns
• .size = Counts the number of individual data points
• .dtype.name = Tells you the data structure type

Below is code that uses all four of these functions with our array.

example.ndim
Out[78]: 2

example.shape
Out[79]: (2, 5)

example.size
Out[80]: 10

example.dtype.name
Out[81]: 'int64'

You can see we have 2 dimensions. The .shape function tells us we have 2 dimensions and 5 examples in each one. The .size function tells us we have 10 total examples (5 * 2). Lastly, the .dtype.name function tells us that this is an integer data type.

Math with Arrays

All mathematical operations can be performed on arrays. Below are examples of addition, subtraction, multiplication, and conditionals.

example=np.array([[1,2,3,4,5],[6,7,8,9,10]]) 
example+2
Out[83]: 
array([[ 3, 4, 5, 6, 7],
[ 8, 9, 10, 11, 12]])

example-2
Out[84]: 
array([[-1, 0, 1, 2, 3],
[ 4, 5, 6, 7, 8]])

example*2
Out[85]: 
array([[ 2, 4, 6, 8, 10],
[12, 14, 16, 18, 20]])

example<3
Out[86]: 
array([[ True, True, False, False, False],
[False, False, False, False, False]], dtype=bool)

Each number inside the example array was manipulated as indicated. For example, if we typed example + 2 all the values in the array increased by 2. Lastly, the example < 3 tells python to look inside the array and find all the values in the array that are less than 3.

Manipulating Arrays

There are also several ways you can manipulate or access data inside an array. For example, you can pull a particular element in an array by doing the following.

example[0,0]
Out[92]: 1

The information in the brackets tells python to access the first bracket and the first number in the bracket. Recall that python starts from 0. You can also access a range of values using the colon as shown below

example=np.array([[1,2,3,4,5],[6,7,8,9,10]]) 
example[:,2:4]
Out[96]: 
array([[3, 4],
[8, 9]])

In this example, the colon means take all values or dimension possible for finding numbers. This means to take columns 1 & 2. After the comma we have 2:4, this means take the 3rd and 4th value but not the 5th.

It is also possible to turn a multidimensional array into a single dimension with the .ravel() function and also to transpose with the transpose() function. Below is the code for each.

example=np.array([[1,2,3,4,5],[6,7,8,9,10]]) 
example.ravel()
Out[97]: array([ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10])

example.transpose()
Out[98]: 
array([[ 1, 6],
[ 2, 7],
[ 3, 8],
[ 4, 9],
[ 5, 10]])

You can see the .ravel function made a one-dimensional array. The .transpose broke the array into several more dimensions with two numbers each.

Conclusion

We now have a basic understanding of how numpy array work using python. As mention before, this is valuable information to understand when trying to wrestling with different data science questions.