gistfile1.txt

# from
# https://www.tensorflow.org/beta/tutorials/estimators/boosted_trees
# TODO: 2.0 is out now
#!pip install tensorflow==2.0.0-rc1
#try:
#    %tensorflow_version 2.x
#except Exception:
#    pass


from __future__ import absolute_import, division, print_function, unicode_literals

import numpy as np
import tensorflow as tf
import platform
import pandas as pd
tf.random.set_seed(123)


print(platform.python_version())
print(tf.version.GIT_VERSION)
print(tf.version.VERSION)
print(tf.__version__)

# Load dataset.
dftrain = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/train.csv')
dfeval = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/eval.csv')
y_train = dftrain.pop('survived')
y_eval = dfeval.pop('survived')

print(dftrain.shape[0])
print(dfeval.shape[0])

fc = tf.feature_column
CATEGORICAL_COLUMNS = ['sex', 'n_siblings_spouses', 'parch', 'class', 'deck', 'embark_town', 'alone']
NUMERIC_COLUMNS = ['age', 'fare']

def one_hot_cat_column(feature_name, vocab):
    return tf.feature_column.indicator_column(
        tf.feature_column.categorical_column_with_vocabulary_list(feature_name, vocab))

feature_columns = []
for feature_name in CATEGORICAL_COLUMNS:
    # Need to one-hot encode categorical features.
    vocabulary = dftrain[feature_name].unique()
    feature_columns.append(one_hot_cat_column(feature_name, vocabulary))

for feature_name in NUMERIC_COLUMNS:
    feature_columns.append(tf.feature_column.numeric_column(feature_name,
                                           dtype=tf.float32))
example = dict(dftrain.head(1))
print(example)
class_fc = tf.feature_column.indicator_column(
    tf.feature_column.categorical_column_with_vocabulary_list(
        'class',
        ('First', 'Second', 'Third')))
print(class_fc)

print('Feature value: \"{}\"'.format(example['class'].iloc[0]))
print('One-hot encoded: ', tf.keras.layers.DenseFeatures([class_fc])(example).numpy())


# Use entire batch since this is such a small dataset.
NUM_EXAMPLES = len(y_train)

def make_input_fn(X, y, n_epochs=None, shuffle=True):
    def input_fn():
        dataset = tf.data.Dataset.from_tensor_slices((dict(X), y))
        if shuffle:
            dataset = dataset.shuffle(NUM_EXAMPLES)
        # For training, cycle thru dataset as many times as need (n_epochs=None).
        dataset = dataset.repeat(n_epochs)
        # In memory training doesn't use batching.
        dataset = dataset.batch(NUM_EXAMPLES)
        return dataset
    return input_fn



# Training and evaluation input functions.
train_input_fn = make_input_fn(dftrain, y_train)
eval_input_fn = make_input_fn(dfeval, y_eval, shuffle=False, n_epochs=1)



# Since data fits into memory, use entire dataset per layer. It will be faster.
# Above one batch is defined as the entire dataset.
n_batches = 1
est = tf.estimator.BoostedTreesClassifier(feature_columns,
                                          n_batches_per_layer=n_batches)

# The model will stop training once the specified number of trees is built, not
# based on the number of steps.
##########
# In jupyter this will crash the kernel
############
est.train(train_input_fn, max_steps=100)

# EVAL
result = est.evalute(eval_input_fn)
print(pd.Series(result))