glaucomunsberg / decision_tree

Implementations of algorithms ID3 and C4.5

Decision Tree

Algorithms ID3 and C4.5 were implemented and the codes run with Python >= 2.7

Run

$ python Main.py [-id3|-c45|-c45p] [-t|-d|-i|-o FILE_PATH]

Attribute	Description
-id3	Execute the ID3 algorithm
-c45	Execute the C45 algorithm
-c45p	Execute the C45 algorithm with pruning
-t FILE	Change the default file to train
-d FILE	Change the default file used on -c45p
-i FILE	File input with examples that you want test
-o FILE	File outut with results from examples tested by the algoritm

Config

You always can change the configuration on the Configuration.py

Attribute	Default value	Description
file_training_url	'data/beach.csv'	File to train
file_depurated_url	'data/beach_depurated.csv'	File with depurate rows to pruning
file_examples_url	'data/examples.csv'	File with examples
file_output_url	'data/output.csv'	File output
splitter	','	Split file when find this caracter
unknown_values	['','-','?']	values that describe a empty value
min_error	1.0	min erro to be use at pruning
print_tabs_num	0	used to format the print method tree
num_split_columns	1	number of columns removed (left side) of the file

Notes

the attributes and values are not sensitive

pruning method only work on C45