Proposal #1 Horse racing dataset.

The dataset is derieved from tips that tipsters provided to bettors.
Tipsters are people that give bettors their best guess aka tip on how to place a bet on a horse race.
This dataset could be used to predict horse races.

Topic:

Interesting Dataset;

Proposal/Yellowbrick:

Classification tutorial with an unbalanced data set

Notes:

This dataset is unbalanced. Given that horse racing is a game of chance, you would expect an unbalanced target.

Other:

  1. Consists of 380k tips
  2. Current features include odds, track, bet-type, horse name, and result
  3. Other features could be added to put together a successful machine learning model such as tempature, humidty, and precipitation.

In [1]:
import pandas as pd
from yellowbrick.classifier.class_balance import ClassBalance
from sklearn.ensemble import RandomForestClassifier
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
from sklearn.model_selection import train_test_split
In [2]:
features = ['Track', 'Horse', 'Bet Type', 'Odds', 'Result']
data = pd.read_csv("https://query.data.world/s/hGxcDSVClhoNX6KacQMHnFVNbHy9Zg",
                   encoding='latin-1')
print(data.shape)
data[features].head()
(38248, 10)
Out[2]:
Track Horse Bet Type Odds Result
0 Ascot Fredricka Win 8.00 Lose
1 Thirsk Spend A Penny Win 4.50 Lose
2 York Straightothepoint Win 7.00 Lose
3 Newmarket Miss Inga Sock Win 5.00 Lose
4 Ascot Peril Win 4.33 Win
In [3]:
# Create training and test sets
X = data.Odds.values.reshape(-1,1)
y = LabelEncoder().fit_transform(data.Result.values.ravel())
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2,
                                                    random_state=42)
classes = data.Result.unique()
In [4]:
# Use Yellowbrick Visualizer ClassBalance
forest = RandomForestClassifier()
visualizer = ClassBalance(forest, classes=classes)

visualizer.fit(X_train, y_train)  # Fit the training data to the visualizer
visualizer.score(X_test, y_test)  # Evaluate the model on the test data
g = visualizer.poof()   
In [ ]:
 

Related content