ICE 3

Link to source code

Team Members:

Name: Avni Mehta, Class Id: 11

Name: (Trinada) Raji Muppala, Class Id: 14

Objective:

Problem 1 :

Build the Linear Regression model using 50 iterations , Split data into 70% training and 30% testing, Report Training Mean Square Error, Report Test Mean Square Error

Input:

Output: Training Mean Square Error, Report Test Mean Square Error

Code Screenshot1

Code Screenshot2

Code Explaination

1. First split the data using "," delimeter
2. Then randomly split the parsed data into 70% Training and 30% test data
3. Use LinearRegressionWithSGD function to train the dataset for 50 iterations and 0.00000001 step size.
4. Then after predicting the values for both training and test dataset, calculate training MSE and test MSE

Problem 2 :

Implement K-Mean Clustering on provided 3D Road Network , Cluster the data into 3 and 4 classes, Evaluate clustering by computing Within Set Sum of Squared Errors , Compare their results

Input:

Output:

Sum of Squared error with Cluster 3

Sum of Squared error with Cluster 4

Comparion chart for culster with 3 cluster with 4

Code

Code Explaination

1. First split the data using "," delimeter
2. Use KMeans function to train the dataset for 3 clusters first.
3. Compute Sum of Squared Errors for 3 clusters and predict values for all clusters.
4. Then use KMeans function to train the dataset for 4 clusters first.
5. Compute Sum of Squared Errors for 4 clusters and predict values for all clusters.

Contritbution of Team Members:

100% for both since both team members completed the ICE individually.