A simple module for using Machine Learning in your code.
Project description
Installation - How to install
The easiest way to install maclearn is using pip. Open your teminal and enter -
pip install maclearn
Instruction - How to use maclearn
Using maclearn in your project is simple enough. Currently, you'll find only 2 algorithms in this module. We are working to develop more.
Reading Excel Files (.csv/.xlsx)
Reading excel files using maclearn is quite easy.Let's say we want to read this file (PizzaPrice.csv) -
Size | Price |
---|---|
6 | 350 |
8 | 775 |
12 | 1150 |
14 | 1395 |
17 | 1648 |
18 | 1675 |
The file may be in .xlsx or .csv format but the extension doesn't matter. Here's how to read it -
For .csv format -
import maclearn # importing the module
filename = "PizzaPrice.csv" # Initialize the directory
file = maclearn.csv(filename) # Create a csv object
data = file.data() # Read the data and store it as a list
print(data) # Print the data
Result - [[6, 8, 12, 14, 17, 18], [350, 775, 1150, 1395, 1648, 1675]]
For .xlsx format -
import maclearn # importing the module filename = "PizzaPrice.xlsx" # Initialize the directory file = maclearn.xlsx(filename) # Create a xlsx object data = file.data() # Read the data and store it as a list print(data) # Print the data
Result - [[6, 8, 12, 14, 17, 18], [350, 775, 1150, 1395, 1648, 1675]]
If you want, you can put file.data()
in the fit method of GaussianNaiveBayes or LinearRegression in order to fit the data to your model from an excel file
Gaussian Naive Bayes Classifier
First create the object GaussianNaiveBayes()...Then call the fit() method to fit the data to your model.
Note: Your data must be in this format - [Feature1, Feature2, Feature3, Label]...
Here, you can use as many features as you want...
After that, simply call the predict(*Features) method to predict the label....
You can calculate the F1 Measure for your model by simply calling the f1_measure() method.
You can also calculate the specificity for your model by simply calling the specificity() method.
Example:
import maclearn # Importing the module
# Features
Outlook = ["Sunny", "Sunny", "Cloudy", "Sunny", "Cloudy", "Cloudy", "Sunny"]
Temperature = ["Cold", "Warm", "Warm", "Warm", "Cold", "Cold", "Cold"]
Routine = ["Indoor", "Outdoor", "Indoor", "Indoor", "Indoor", "Outdoor", "Outdoor", "Outdoor"]
WearCoat = ["No", "No", "No", "No", "Yes", "Yes", "Yes"] # Labels
# You can also import the data from an excel file
Features = [Outlook, Temperature, Routine, WearCoat] # Putting all features in a list
model = maclearn.GaussianNaiveBayes() # Creating the object
model.fit(Features) # Fit the data into the model
# Should we wear coat, if our feature set is cloudy, warm and outdoor?
print(model.predict("Cloudy", "Warm", "Outdoor")) # Predict the label
f1_value = model.f1_measure() # Calculate the F1 measure value for the model
specificity = model.specificity() # Calculate the specificity for the model
print(f1_value * 100, "%") # print F1 measure value in percentage
print(specificity * 100, "%") # print specificity in percentage
Linear Regression with numerical analysis
First create the object - LinearRegression()...Then call the fit(x, y) method to fit the data to your model.
This model is for single variable linear regression. So your data must be in this format - X=[2,3,4], Y=[5,6,7]
Then, just call the predict(x) method with the argument of x value to predict the y value...
You can also calculate the R-Squared value by simply calling r_squared() method.
Example:
import maclearn # Importing the module
# You can also import the data from an excel file
Size = [6, 8, 12, 14, 18] # Size of Pizza in inches (Independent Variable)
Price = [350, 775, 1150, 1395, 1675] # Price of Pizza in Taka (Dependent Variable)
model = maclearn.LinearRegression() # Creating the object
model.fit(Size, Price) # Fit the data into the model
predicted = model.predict(17) # Predicted the price of 17 inches pizza
accuracy = model.r_squared() # Calculate the R-Squared value for the model
print(predicted) # Print the predicted value
print(accuracy * 100, "%") # Print the R-Squared value in percenntage
Lisence
This module is completely free and open source. You can use and modify to improve the module if you want ;)
Any suggestion will be highly appriciated. Gmail -
neural.gen.official@gmail.com
Created by Sajedur Rahman Fiad
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
MacLearn-1.0.0.tar.gz
(5.3 kB
view hashes)