Titanic dataset - Categorical example

titanic_categorical1.py

import random
import pandas
import numpy as np
from sklearn import metrics, cross_validation

import tensorflow as tf
from tensorflow.contrib import layers
from tensorflow.contrib import learn

random.seed(42)

data = pandas.read_csv('data/titanic_train.csv')
X = data[["Embarked"]]
y = data["Survived"]
X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.2, random_state=42)

embarked_classes = X_train["Embarked"].unique()
n_classes = len(embarked_classes) + 1
print('Embarked has next classes: ', embarked_classes)

cat_processor = learn.preprocessing.CategoricalProcessor()
X_train = np.array(list(cat_processor.fit_transform(X_train)))
X_test = np.array(list(cat_processor.transform(X_test)))

### Embeddings

EMBEDDING_SIZE = 3

def categorical_model(features, target):
    target = tf.one_hot(target, 2, 1.0, 0.0)
    features = learn.ops.categorical_variable(
        features, n_classes, embedding_size=EMBEDDING_SIZE, name='embarked')
    prediction, loss = learn.models.logistic_regression(tf.squeeze(features, [1]), target)
    train_op = layers.optimize_loss(loss,
        tf.contrib.framework.get_global_step(), optimizer='SGD', learning_rate=0.05)
    return tf.argmax(prediction, dimension=1), loss, train_op

classifier = learn.Estimator(model_fn=categorical_model)
classifier.fit(X_train, y_train, steps=1000)

print("Accuracy: {0}".format(metrics.accuracy_score(classifier.predict(X_test), y_test)))
print("ROC: {0}".format(metrics.roc_auc_score(classifier.predict(X_test), y_test)))

### One Hot

def one_hot_categorical_model(features, target):
    target = tf.one_hot(target, 2, 1.0, 0.0)
    features = tf.one_hot(features, n_classes, 1.0, 0.0)
    prediction, loss = learn.models.logistic_regression(
      tf.squeeze(features, [1]), target)
    train_op = layers.optimize_loss(loss,
        tf.contrib.framework.get_global_step(), optimizer='SGD',
        learning_rate=0.01)
    return tf.argmax(prediction, dimension=1), loss, train_op

classifier = learn.Estimator(model_fn=one_hot_categorical_model)
classifier.fit(X_train, y_train, steps=1000)

print("Accuracy: {0}".format(metrics.accuracy_score(classifier.predict(X_test), y_test)))
print("ROC: {0}".format(metrics.roc_auc_score(classifier.predict(X_test), y_test)))