Titanic Survival Prediction

Search for a good model for the Titanic dataset. First, you need to download the titanic dataset file train.csv and eval.csv. Second, replace PATH_TO/train.csv and PATH_TO/eval.csv in the following example with the real path to those two files. Then, you can run the code.

import autokeras as ak

# Initialize the classifier.
clf = ak.StructuredDataClassifier(max_trials=30)
# x is the path to the csv file. y is the column name of the column to predict.
clf.fit(x='PATH_TO/train.csv', y='survived')
# Evaluate the accuracy of the found model.
print('Accuracy: {accuracy}'.format(
    accuracy=clf.evaluate(x='PATH_TO/eval.csv', y='survived')))