TL;DR, predicting poker hands.

Given the sequence of 5 'community' cards, drawn from a standard deck of cards, what hand is most likely present with at least one of the players in the game.

Diligence is the mother of good luck. 
                            - Benjamin Franklin

Texas Hold Em is played by dealing each player 2 cards (face down), called the hole cards, and dealing 5 community cards (face up), on the table.

The player makes a poker hand using any combination of the 3 cards dealt to them, and the 5 cards on the table.

The player with the strongest hands wins. Most commonly accepted ranking of hands, strongest to weekest :

A K Q J 10 9 8 7 6 5 4 3 2 1

This data was acquired from UCI's Machine Learning repository. The data comes already split into training and testing data.

Find it here.

Each bar represents the frequency of the corresponding hand as specified in the dataset description.

Dua, D. and Karra Taniskidou, E. (2017). UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science.

Model	Accuracy	Plot
Linear Regression	42%
SVM	58%
Adaboost	49%
Output Code Classifier	61%
Random Forest	56%
Artificial Neural Network	45%
Deep Neural Network	87%
Multi Layer Perceptron	97% 🤯

The Multi-layer Perceptron is clearly the best model for the dataset in hand. Here's the confusion matrix to provide us with some more insight into how accurate this model really was with its predictions.

	Aditi Srinivas
	Avinash Shenoy