mjoshy2/cluster_script

Class Decomposition via Clustering Research Project

Class Decomposition via Clustering Project

The project can be run by simply executing the build script.

>>> sh build.sh

Simply cd into the root directory and run the above command. Note: If you use windows you need to have bash installed.

In the future we will be using more datasets. For that purpose the directory is set up such that all the excel and csv files go in the data folder. The scripts to clean the data and feed it to the neural net are in the src folder. And the functions needed to manage dataframe objects go in the utility folder.

cluster_script's Issues

Missing parameter

TypeError: pre_process_data() missing 1 required positional argument: 'target_col'

As I see you are not working on categorial data so there is an extra parameter here that is not being used. I removed it.

pre_process_data(dataframe, numerical_col, categorical_col, target_col):

Missing function from Utilities

get_accuracy is missing. I will write something of my own in a bit.

Extra parameter not used but only assigned?

File "credit_data_clusterednn.py", line 48, in <module>
    x_train, y_train, x_test, y_test, test_true_labels = Util.split_data(0.8, train_x, train_y)
ValueError: not enough values to unpack (expected 5, got 4)

Changed to:

x_train, y_train, x_test, y_test  = Util.split_data(0.8, train_x, train_y)

test_true_labels = true_labels