CS5350/6350-Machine-Learning

This is a machine learning library developed by Mokshagna Sai Teja Karanam for CS5350/6350 in University of Utah.

For running the Decision Tree

Installation instructions(Run this before you run the other shell-scripts):

# Installs pandas, numpy
./install.sh

Commands to run the shell-scripts

# To run various cases question-2a:
# Please use the same strings for metric names: gini, MajorityError, entropy
Examples:
./run_2a.sh 1 gini
./run_2a.sh 3 MajorityError
./run_2a.sh 5 entropy

For running the Ensemble Learning, Linear Regression

Before running the sh files, please make sure permissions are enabled:

Installation instructions(Run this before you run the other shell-scripts):

# Install pandas, numpy, pickle, matplotlib, xlrd(for bonus) version > 1.0.0
./install.sh

For running the Perceptron

Before running the sh files, please make sure permissions are enabled by executing the following commands:

chmod +x install.sh

chmod +x run_2a.sh

chmod +x run_2b.sh

chmod +x run_2c.sh

Installation instructions(Run this before you run the other shell-scripts):

# Install pandas, numpy by running
./install.sh

For running the SVM

Run the sh files for each problem like run_2a.sh and so on:

Installation instructions:

# Install pandas, numpy, scipy, warnings by running
./install.sh

For running the Neural Network

Run the sh files for each problem to get the results:

Before running the sh files, please make sure permissions are enabled by executing the following commands:

chmod +x install.sh

chmod +x run_2a1.sh

chmod +x run_2a2.sh

chmod +x run_2b.sh

chmod +x run_2c.sh

chmod +x run_2e.sh

Installation instructions:

# Install pandas, numpy, tensorflow, warnings by running
./install.sh

I have created 2 notebooks for 2a question where 2a1 includes running one training example of bank dataset and it will list out all the gradients of weights and it will take 2 arguments represents width of each layer as this question is asking 2 layers give 2 arguments as follows to see the results:

run_2a1.sh 5 5

run_2a1.sh 2 4

#####For running 2a2 (verify the results of problem 3) run the command:

run_2a2.sh

For running the Logistic Regression

Before running the sh files, please make sure permissions are enabled by executing the following commands:

chmod +x install.sh

chmod +x run_3a.sh

chmod +x run_3b.sh

Installation instructions(Run this before you run the other shell-scripts):

# Install pandas, numpy by running
./install.sh

For running the ML Project

In this github repository the project is uploaded in the folder ML Project where make sure the install.sh and run project.sh by running the following:

chmod +x install.sh

chmod +x run project.sh

Installation instructions(Run this before you run the shell-script):

./install.sh

By running the project (run project.sh) all the models that is discussed above will invoke and save the respective predicted test probability values in the same path.