Lab Assignment 4 - Achu0015/BigDataAna-App-SP2K17 GitHub Wiki

                                             # Lab Assignment 4

Image classification is done on set of data of Integers and letters. The digits 0, 1, 4 and letter A are trained using spark MLlib library.

I have collected a data set of images of '0', '1', '4' and 'A', all these images are trained and another data set of images for testing are used.

Using Decision Tree model algorithm we generate the histograms and confusion matrix for the data set, then finally the accuracy is measured on the scale of 0 to 1.

Below are the screen shots of procedure:

The screen shots of the Test Data are below:

The screen shots of Train data are below:

Using the Spark MLlib library Decision Tree model was used here are the code snippets:

Once the spark program executed successfully, we generate the key descriptors, Histograms, Confusion Matrix and Accuracy score: