Basmodell

Vinna en Kaggle-tävling i Python

Yauhen Babakhin

Kaggle Grandmaster

Modelleringsfasen

komponenter i modelleringsfasen

Modelleringsfasen

basmodell i modelleringsfasen

Lokal validering för New York-taxi

# Read data
taxi_train = pd.read_csv('taxi_train.csv')
taxi_test = pd.read_csv('taxi_test.csv')

from sklearn.model_selection import train_test_split

# Create local validation
validation_train, validation_test = train_test_split(taxi_train,
                                                     test_size=0.3,
                                                     random_state=123)

Basmodell I

import numpy as np
# Assign the mean fare amount to all the test observations
taxi_test['fare_amount'] = np.mean(taxi_train.fare_amount)

# Write predictions to the file
taxi_test[['id','fare_amount']].to_csv('mean_sub.csv', index=False)

Validerings-RMSE	Public LB RMSE	Public LB-position
9.986	9.409	1449 / 1500

Basmodell II

# Calculate the mean fare amount by group
naive_prediction_groups = taxi_train.groupby('passenger_count').fare_amount.mean()

# Make predictions on the test set
taxi_test['fare_amount'] = taxi_test.passenger_count.map(naive_prediction_groups)

# Write predictions to the file
taxi_test[['id','fare_amount']].to_csv('mean_group_sub.csv', index=False)

Validerings-RMSE	Public LB RMSE	Public LB-position
9.978	9.407	1411 / 1500

Basmodell III

# Select only numeric features
features = ['pickup_longitude', 'pickup_latitude',
            'dropoff_longitude', 'dropoff_latitude', 'passenger_count']

from sklearn.ensemble import GradientBoostingRegressor

# Train a Gradient Boosting model
gb = GradientBoostingRegressor()
gb.fit(taxi_train[features], taxi_train.fare_amount)

# Make predictions on the test data
taxi_test['fare_amount'] = gb.predict(taxi_test[features])

Basmodell III

# Write predictions to the file
taxi_test[['id','fare_amount']].to_csv('gb_sub.csv', index=False)

Validerings-RMSE	Public LB RMSE	Public LB-position
5.996	4.595	1109 / 1500

Preliminära resultat

Modell	Validerings-RMSE	Public LB RMSE
Enkelt medelvärde	9.986	9.409
Gruppmedelvärde	9.978	9.407
Gradient Boosting	5.996	4.595

Korrelation med public leaderboard

Modell	Validerings-RMSE	Public LB RMSE
Modell A	3.500	3.800
Modell B	3.300	4.100
Modell C	3.200	3.900

Modell	Validerings-RMSE	Public LB RMSE
Modell A	3.400	3.900
Modell B	3.100	3.400
Modell C	2.900	3.300

Nu kör vi en övning!

Vinna en Kaggle-tävling i Python