kunal-visoulia / gender_classification_based_on_shoesize_and_height

Multiple classifiers and their accuaracy comparison

Gender Classification by ShoeSize and Height

Siraj Raval's 'Learn Python for Data science' playlist Video 1 https://youtu.be/T5pRlIbr6gg?list=PL2-dafEMk2A6QKz1mrk1uIGfHkC1zZ6UU

Note: I used a different and larger dataset available on internet for training and testing the models.

http://www.academia.edu/29485664/Correlation_Between_Gender_Height_And_Shoe_Size_In_Australian_University_Students

Challenge:

Find 3 more classifiers from the sci-kit learn documentation and add them to the code. Train them on the same dataset and compare their results. The following classifiers were chosen:

• Decision Tree Classifier

• Nearest Neighbour Classifier

• Random Forest Classifier

• Naive Bayes Classifier

Usage:

just run python classifier.py

Status:

• DecisionTreeClassifier    Accuracy:  90.98%
• KNeighborsClassifier      Accuracy:  87.27%
• RandomForestClassifier    Accuracy:  90.72%
• GaussianNB             	Accuracy:  84.88%